cephi/ampere_research
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

as-caida max

Browse Source
This commit is contained in:
cephi 2024-12-15 17:34:46 -05:00
parent e6cd4fb2c0
commit d5caf9c543
144 changed files with 5870 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_005.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 11868, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 20.50465679168701, "TIME_S_1KI": 1.7277263895927715, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 595.5481514930726, "W": 24.482217423939822, "J_1KI": 50.1810036647348, "W_1KI": 2.062876426014478, "W_D": 6.179217423939821, "J_D": 150.3140606415273, "W_D_1KI": 0.5206620680771672, "J_D_1KI": 0.04387108763710543}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_005.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_005.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 1.769423484802246}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.9190, 0.7017, 0.0047, ..., 0.0989, 0.5520, 0.9034])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 1.769423484802246 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 11868 -m matrices/as-caida_pruned/as-caida_G_005.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 20.50465679168701}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.1433, 0.5564, 0.9502, ..., 0.1610, 0.5318, 0.0201])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 20.50465679168701 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.1433, 0.5564, 0.9502, ..., 0.1610, 0.5318, 0.0201])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 20.50465679168701 seconds
[20.2, 20.44, 20.4, 20.48, 20.44, 20.52, 20.32, 20.4, 20.68, 20.68]
[20.8, 20.6, 20.92, 25.16, 26.4, 33.24, 34.12, 31.92, 30.84, 24.28, 24.16, 24.2, 24.2, 24.16, 24.04, 24.16, 24.48, 24.6, 24.52, 24.8, 24.84, 24.8, 24.56, 24.4]
24.325743913650513
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 11868, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_005', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 70026, 'MATRIX_DENSITY': 7.111825976492498e-05, 'TIME_S': 20.50465679168701, 'TIME_S_1KI': 1.7277263895927715, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 595.5481514930726, 'W': 24.482217423939822}
[20.2, 20.44, 20.4, 20.48, 20.44, 20.52, 20.32, 20.4, 20.68, 20.68, 20.04, 19.88, 19.88, 19.88, 20.12, 20.4, 20.64, 20.36, 20.52, 20.48]
366.06
18.303
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 11868, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_005', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 70026, 'MATRIX_DENSITY': 7.111825976492498e-05, 'TIME_S': 20.50465679168701, 'TIME_S_1KI': 1.7277263895927715, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 595.5481514930726, 'W': 24.482217423939822, 'J_1KI': 50.1810036647348, 'W_1KI': 2.062876426014478, 'W_D': 6.179217423939821, 'J_D': 150.3140606415273, 'W_D_1KI': 0.5206620680771672, 'J_D_1KI': 0.04387108763710543}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_010.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 11106, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 20.421429872512817, "TIME_S_1KI": 1.838774524807565, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 604.79751707077, "W": 23.847910243049302, "J_1KI": 54.45682667664055, "W_1KI": 2.1472996797271113, "W_D": 5.517910243049304, "J_D": 139.93756184101088, "W_D_1KI": 0.49684046848994273, "J_D_1KI": 0.044736220825674654}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_010.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_010.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 1.8907017707824707}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.9147, 0.8941, 0.7248, ..., 0.0711, 0.2516, 0.9374])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 1.8907017707824707 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 11106 -m matrices/as-caida_pruned/as-caida_G_010.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 20.421429872512817}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.3281, 0.7131, 0.6291, ..., 0.9063, 0.8724, 0.7956])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 20.421429872512817 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.3281, 0.7131, 0.6291, ..., 0.9063, 0.8724, 0.7956])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 20.421429872512817 seconds
[20.4, 20.28, 20.36, 20.52, 20.48, 20.6, 20.56, 20.36, 20.32, 20.12]
[20.04, 20.32, 20.32, 21.2, 22.36, 26.6, 27.8, 28.68, 28.68, 28.44, 25.24, 24.84, 24.8, 25.08, 25.08, 24.84, 24.96, 24.96, 24.72, 24.84, 25.12, 24.92, 24.84, 24.76, 24.84]
25.36060857772827
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 11106, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_010', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 74994, 'MATRIX_DENSITY': 7.616375021864427e-05, 'TIME_S': 20.421429872512817, 'TIME_S_1KI': 1.838774524807565, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 604.79751707077, 'W': 23.847910243049302}
[20.4, 20.28, 20.36, 20.52, 20.48, 20.6, 20.56, 20.36, 20.32, 20.12, 20.28, 20.4, 20.32, 20.32, 20.4, 20.4, 20.56, 20.28, 20.0, 20.08]
366.59999999999997
18.33
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 11106, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_010', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 74994, 'MATRIX_DENSITY': 7.616375021864427e-05, 'TIME_S': 20.421429872512817, 'TIME_S_1KI': 1.838774524807565, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 604.79751707077, 'W': 23.847910243049302, 'J_1KI': 54.45682667664055, 'W_1KI': 2.1472996797271113, 'W_D': 5.517910243049304, 'J_D': 139.93756184101088, 'W_D_1KI': 0.49684046848994273, 'J_D_1KI': 0.044736220825674654}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_015.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 10628, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_015", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 77124, "MATRIX_DENSITY": 7.832697378273889e-05, "TIME_S": 20.489109754562378, "TIME_S_1KI": 1.9278424684383118, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 592.0631146240235, "W": 24.361970542226484, "J_1KI": 55.707857981183984, "W_1KI": 2.292244123280625, "W_D": 5.985970542226486, "J_D": 145.47560334396368, "W_D_1KI": 0.5632264341575542, "J_D_1KI": 0.05299458356770364}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_015.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_015.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_015", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 77124, "MATRIX_DENSITY": 7.832697378273889e-05, "TIME_S": 1.9757425785064697}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 4, 4, ..., 77124, 77124, 77124]),
col_indices=tensor([1040, 2054, 4842, ..., 160, 160, 8230]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=77124, layout=torch.sparse_csr)
tensor([0.7519, 0.6257, 0.7169, ..., 0.6184, 0.4198, 0.1096])
Matrix Type: SuiteSparse
Matrix: as-caida_G_015
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 77124
Density: 7.832697378273889e-05
Time: 1.9757425785064697 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 10628 -m matrices/as-caida_pruned/as-caida_G_015.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_015", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 77124, "MATRIX_DENSITY": 7.832697378273889e-05, "TIME_S": 20.489109754562378}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 4, 4, ..., 77124, 77124, 77124]),
col_indices=tensor([1040, 2054, 4842, ..., 160, 160, 8230]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=77124, layout=torch.sparse_csr)
tensor([0.7931, 0.3419, 0.7780, ..., 0.8001, 0.8539, 0.6736])
Matrix Type: SuiteSparse
Matrix: as-caida_G_015
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 77124
Density: 7.832697378273889e-05
Time: 20.489109754562378 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 4, 4, ..., 77124, 77124, 77124]),
col_indices=tensor([1040, 2054, 4842, ..., 160, 160, 8230]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=77124, layout=torch.sparse_csr)
tensor([0.7931, 0.3419, 0.7780, ..., 0.8001, 0.8539, 0.6736])
Matrix Type: SuiteSparse
Matrix: as-caida_G_015
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 77124
Density: 7.832697378273889e-05
Time: 20.489109754562378 seconds
[20.48, 20.4, 20.32, 20.36, 20.68, 20.4, 20.36, 20.24, 20.32, 20.2]
[20.2, 20.32, 20.56, 24.04, 25.84, 32.64, 33.32, 33.6, 30.12, 26.64, 24.0, 23.92, 24.12, 24.12, 24.32, 24.28, 24.44, 24.56, 24.24, 24.52, 24.44, 24.2, 24.2, 24.32]
24.302759647369385
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 10628, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_015', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 77124, 'MATRIX_DENSITY': 7.832697378273889e-05, 'TIME_S': 20.489109754562378, 'TIME_S_1KI': 1.9278424684383118, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 592.0631146240235, 'W': 24.361970542226484}
[20.48, 20.4, 20.32, 20.36, 20.68, 20.4, 20.36, 20.24, 20.32, 20.2, 20.52, 20.44, 20.44, 20.2, 20.36, 20.4, 20.36, 20.52, 20.8, 20.64]
367.52
18.375999999999998
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 10628, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_015', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 77124, 'MATRIX_DENSITY': 7.832697378273889e-05, 'TIME_S': 20.489109754562378, 'TIME_S_1KI': 1.9278424684383118, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 592.0631146240235, 'W': 24.361970542226484, 'J_1KI': 55.707857981183984, 'W_1KI': 2.292244123280625, 'W_D': 5.985970542226486, 'J_D': 145.47560334396368, 'W_D_1KI': 0.5632264341575542, 'J_D_1KI': 0.05299458356770364}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_020.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 10071, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_020", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 80948, "MATRIX_DENSITY": 8.221062021893506e-05, "TIME_S": 20.918866634368896, "TIME_S_1KI": 2.077138976702303, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 573.2929569816588, "W": 24.61671638273014, "J_1KI": 56.925127294375805, "W_1KI": 2.4443169876606237, "W_D": 6.3057163827301395, "J_D": 146.85235572195043, "W_D_1KI": 0.626126142660127, "J_D_1KI": 0.062171198754853246}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_020.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_020.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_020", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 80948, "MATRIX_DENSITY": 8.221062021893506e-05, "TIME_S": 2.085054636001587}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 80944, 80946, 80948]),
col_indices=tensor([ 1040, 5699, 106, ..., 31378, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=80948, layout=torch.sparse_csr)
tensor([0.5085, 0.4122, 0.0237, ..., 0.5225, 0.0817, 0.0530])
Matrix Type: SuiteSparse
Matrix: as-caida_G_020
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 80948
Density: 8.221062021893506e-05
Time: 2.085054636001587 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 10071 -m matrices/as-caida_pruned/as-caida_G_020.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_020", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 80948, "MATRIX_DENSITY": 8.221062021893506e-05, "TIME_S": 20.918866634368896}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 80944, 80946, 80948]),
col_indices=tensor([ 1040, 5699, 106, ..., 31378, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=80948, layout=torch.sparse_csr)
tensor([0.4326, 0.5436, 0.4160, ..., 0.2480, 0.2557, 0.8019])
Matrix Type: SuiteSparse
Matrix: as-caida_G_020
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 80948
Density: 8.221062021893506e-05
Time: 20.918866634368896 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 80944, 80946, 80948]),
col_indices=tensor([ 1040, 5699, 106, ..., 31378, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=80948, layout=torch.sparse_csr)
tensor([0.4326, 0.5436, 0.4160, ..., 0.2480, 0.2557, 0.8019])
Matrix Type: SuiteSparse
Matrix: as-caida_G_020
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 80948
Density: 8.221062021893506e-05
Time: 20.918866634368896 seconds
[20.52, 20.32, 20.24, 20.16, 20.0, 20.04, 20.32, 20.44, 20.6, 20.44]
[20.6, 20.44, 20.44, 23.72, 25.68, 32.48, 33.28, 33.84, 30.04, 26.88, 24.32, 24.48, 24.44, 24.56, 24.56, 24.8, 24.92, 25.0, 24.96, 24.84, 25.0, 24.72, 24.84]
23.288766384124756
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 10071, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_020', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 80948, 'MATRIX_DENSITY': 8.221062021893506e-05, 'TIME_S': 20.918866634368896, 'TIME_S_1KI': 2.077138976702303, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 573.2929569816588, 'W': 24.61671638273014}
[20.52, 20.32, 20.24, 20.16, 20.0, 20.04, 20.32, 20.44, 20.6, 20.44, 20.32, 20.36, 20.28, 20.44, 20.48, 20.48, 20.48, 20.4, 20.32, 20.44]
366.22
18.311
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 10071, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_020', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 80948, 'MATRIX_DENSITY': 8.221062021893506e-05, 'TIME_S': 20.918866634368896, 'TIME_S_1KI': 2.077138976702303, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 573.2929569816588, 'W': 24.61671638273014, 'J_1KI': 56.925127294375805, 'W_1KI': 2.4443169876606237, 'W_D': 6.3057163827301395, 'J_D': 146.85235572195043, 'W_D_1KI': 0.626126142660127, 'J_D_1KI': 0.062171198754853246}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_025.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 9843, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_025", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 85850, "MATRIX_DENSITY": 8.718908121010495e-05, "TIME_S": 20.97947359085083, "TIME_S_1KI": 2.1314105039978495, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 582.6910747146605, "W": 23.96309690957501, "J_1KI": 59.198524303023525, "W_1KI": 2.434531840858987, "W_D": 5.643096909575011, "J_D": 137.21858303070056, "W_D_1KI": 0.5733106684522007, "J_D_1KI": 0.058245521533292766}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_025.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_025.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_025", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 85850, "MATRIX_DENSITY": 8.718908121010495e-05, "TIME_S": 2.1334664821624756}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 3, 3, ..., 85845, 85847, 85850]),
col_indices=tensor([ 346, 13811, 21783, ..., 15310, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=85850, layout=torch.sparse_csr)
tensor([0.1155, 0.2832, 0.4927, ..., 0.9506, 0.2786, 0.8718])
Matrix Type: SuiteSparse
Matrix: as-caida_G_025
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 85850
Density: 8.718908121010495e-05
Time: 2.1334664821624756 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 9843 -m matrices/as-caida_pruned/as-caida_G_025.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_025", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 85850, "MATRIX_DENSITY": 8.718908121010495e-05, "TIME_S": 20.97947359085083}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 3, 3, ..., 85845, 85847, 85850]),
col_indices=tensor([ 346, 13811, 21783, ..., 15310, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=85850, layout=torch.sparse_csr)
tensor([0.9708, 0.2572, 0.8092, ..., 0.0249, 0.0177, 0.4715])
Matrix Type: SuiteSparse
Matrix: as-caida_G_025
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 85850
Density: 8.718908121010495e-05
Time: 20.97947359085083 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 3, 3, ..., 85845, 85847, 85850]),
col_indices=tensor([ 346, 13811, 21783, ..., 15310, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=85850, layout=torch.sparse_csr)
tensor([0.9708, 0.2572, 0.8092, ..., 0.0249, 0.0177, 0.4715])
Matrix Type: SuiteSparse
Matrix: as-caida_G_025
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 85850
Density: 8.718908121010495e-05
Time: 20.97947359085083 seconds
[19.8, 20.16, 20.16, 20.24, 20.4, 20.36, 20.44, 20.44, 20.6, 20.68]
[20.56, 20.52, 20.4, 22.8, 23.4, 29.48, 30.2, 30.32, 29.48, 24.36, 24.36, 24.52, 24.88, 24.8, 24.72, 24.76, 24.6, 24.2, 24.32, 24.44, 24.76, 24.96, 25.4, 25.4]
24.316184043884277
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9843, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_025', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 85850, 'MATRIX_DENSITY': 8.718908121010495e-05, 'TIME_S': 20.97947359085083, 'TIME_S_1KI': 2.1314105039978495, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 582.6910747146605, 'W': 23.96309690957501}
[19.8, 20.16, 20.16, 20.24, 20.4, 20.36, 20.44, 20.44, 20.6, 20.68, 20.36, 20.48, 20.6, 20.4, 20.52, 20.6, 20.4, 20.0, 20.12, 20.12]
366.40000000000003
18.32
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9843, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_025', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 85850, 'MATRIX_DENSITY': 8.718908121010495e-05, 'TIME_S': 20.97947359085083, 'TIME_S_1KI': 2.1314105039978495, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 582.6910747146605, 'W': 23.96309690957501, 'J_1KI': 59.198524303023525, 'W_1KI': 2.434531840858987, 'W_D': 5.643096909575011, 'J_D': 137.21858303070056, 'W_D_1KI': 0.5733106684522007, 'J_D_1KI': 0.058245521533292766}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_030.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 9604, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_030", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 86850, "MATRIX_DENSITY": 8.820467912752026e-05, "TIME_S": 20.284237384796143, "TIME_S_1KI": 2.1120613686793153, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 606.7543531799317, "W": 24.93372016805261, "J_1KI": 63.17725460015949, "W_1KI": 2.5961807755156823, "W_D": 6.76272016805261, "J_D": 164.56869948196425, "W_D_1KI": 0.704156618914266, "J_D_1KI": 0.07331909817932798}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_030.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_030.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_030", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 86850, "MATRIX_DENSITY": 8.820467912752026e-05, "TIME_S": 2.1865053176879883}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 86850, 86850, 86850]),
col_indices=tensor([ 1809, 21783, 106, ..., 7018, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=86850, layout=torch.sparse_csr)
tensor([0.6109, 0.9417, 0.2480, ..., 0.3925, 0.8143, 0.3245])
Matrix Type: SuiteSparse
Matrix: as-caida_G_030
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 86850
Density: 8.820467912752026e-05
Time: 2.1865053176879883 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 9604 -m matrices/as-caida_pruned/as-caida_G_030.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_030", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 86850, "MATRIX_DENSITY": 8.820467912752026e-05, "TIME_S": 20.284237384796143}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 86850, 86850, 86850]),
col_indices=tensor([ 1809, 21783, 106, ..., 7018, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=86850, layout=torch.sparse_csr)
tensor([0.8685, 0.7727, 0.7628, ..., 0.3143, 0.3501, 0.5646])
Matrix Type: SuiteSparse
Matrix: as-caida_G_030
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 86850
Density: 8.820467912752026e-05
Time: 20.284237384796143 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 86850, 86850, 86850]),
col_indices=tensor([ 1809, 21783, 106, ..., 7018, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=86850, layout=torch.sparse_csr)
tensor([0.8685, 0.7727, 0.7628, ..., 0.3143, 0.3501, 0.5646])
Matrix Type: SuiteSparse
Matrix: as-caida_G_030
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 86850
Density: 8.820467912752026e-05
Time: 20.284237384796143 seconds
[20.04, 19.8, 19.96, 20.12, 20.08, 20.32, 20.4, 20.6, 20.48, 20.4]
[20.2, 20.28, 23.2, 23.2, 24.64, 31.64, 32.88, 33.36, 30.64, 30.28, 25.0, 25.32, 25.12, 25.16, 24.88, 24.88, 24.76, 24.96, 25.04, 24.84, 25.04, 25.16, 25.24, 25.32]
24.33469009399414
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9604, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_030', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 86850, 'MATRIX_DENSITY': 8.820467912752026e-05, 'TIME_S': 20.284237384796143, 'TIME_S_1KI': 2.1120613686793153, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 606.7543531799317, 'W': 24.93372016805261}
[20.04, 19.8, 19.96, 20.12, 20.08, 20.32, 20.4, 20.6, 20.48, 20.4, 20.16, 20.0, 20.12, 20.04, 20.36, 20.36, 20.04, 20.0, 20.2, 20.48]
363.41999999999996
18.171
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9604, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_030', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 86850, 'MATRIX_DENSITY': 8.820467912752026e-05, 'TIME_S': 20.284237384796143, 'TIME_S_1KI': 2.1120613686793153, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 606.7543531799317, 'W': 24.93372016805261, 'J_1KI': 63.17725460015949, 'W_1KI': 2.5961807755156823, 'W_D': 6.76272016805261, 'J_D': 164.56869948196425, 'W_D_1KI': 0.704156618914266, 'J_D_1KI': 0.07331909817932798}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_035.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 9472, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_035", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 87560, "MATRIX_DENSITY": 8.892575364888514e-05, "TIME_S": 20.188151359558105, "TIME_S_1KI": 2.1313504391425364, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 612.7760334014894, "W": 25.19675156600016, "J_1KI": 64.69341568850183, "W_1KI": 2.6601300217483277, "W_D": 6.96475156600016, "J_D": 169.3802801151277, "W_D_1KI": 0.7352989406672467, "J_D_1KI": 0.07762868883733601}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_035.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_035.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_035", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 87560, "MATRIX_DENSITY": 8.892575364888514e-05, "TIME_S": 2.2169973850250244}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 87559, 87559, 87560]),
col_indices=tensor([ 1809, 21783, 106, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=87560, layout=torch.sparse_csr)
tensor([0.2444, 0.3577, 0.2903, ..., 0.6956, 0.4519, 0.7042])
Matrix Type: SuiteSparse
Matrix: as-caida_G_035
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 87560
Density: 8.892575364888514e-05
Time: 2.2169973850250244 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 9472 -m matrices/as-caida_pruned/as-caida_G_035.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_035", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 87560, "MATRIX_DENSITY": 8.892575364888514e-05, "TIME_S": 20.188151359558105}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 87559, 87559, 87560]),
col_indices=tensor([ 1809, 21783, 106, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=87560, layout=torch.sparse_csr)
tensor([0.3089, 0.5002, 0.8484, ..., 0.9388, 0.8606, 0.9713])
Matrix Type: SuiteSparse
Matrix: as-caida_G_035
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 87560
Density: 8.892575364888514e-05
Time: 20.188151359558105 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 87559, 87559, 87560]),
col_indices=tensor([ 1809, 21783, 106, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=87560, layout=torch.sparse_csr)
tensor([0.3089, 0.5002, 0.8484, ..., 0.9388, 0.8606, 0.9713])
Matrix Type: SuiteSparse
Matrix: as-caida_G_035
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 87560
Density: 8.892575364888514e-05
Time: 20.188151359558105 seconds
[20.32, 20.32, 20.32, 20.2, 20.04, 20.04, 19.88, 20.24, 20.36, 20.28]
[20.68, 20.6, 20.84, 25.76, 30.24, 33.32, 34.32, 31.76, 31.76, 30.56, 24.76, 24.52, 24.48, 24.48, 24.52, 25.2, 25.36, 25.48, 25.36, 24.84, 24.84, 24.6, 24.64, 24.64]
24.31964421272278
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9472, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_035', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 87560, 'MATRIX_DENSITY': 8.892575364888514e-05, 'TIME_S': 20.188151359558105, 'TIME_S_1KI': 2.1313504391425364, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 612.7760334014894, 'W': 25.19675156600016}
[20.32, 20.32, 20.32, 20.2, 20.04, 20.04, 19.88, 20.24, 20.36, 20.28, 20.4, 20.36, 20.28, 20.16, 20.12, 20.4, 20.56, 20.44, 20.28, 20.28]
364.64
18.232
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9472, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_035', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 87560, 'MATRIX_DENSITY': 8.892575364888514e-05, 'TIME_S': 20.188151359558105, 'TIME_S_1KI': 2.1313504391425364, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 612.7760334014894, 'W': 25.19675156600016, 'J_1KI': 64.69341568850183, 'W_1KI': 2.6601300217483277, 'W_D': 6.96475156600016, 'J_D': 169.3802801151277, 'W_D_1KI': 0.7352989406672467, 'J_D_1KI': 0.07762868883733601}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_040.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 9338, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_040", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89658, "MATRIX_DENSITY": 9.105647807962247e-05, "TIME_S": 20.749686002731323, "TIME_S_1KI": 2.222069608345612, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 604.0135382080078, "W": 24.829919965337524, "J_1KI": 64.68339453930263, "W_1KI": 2.659019058185642, "W_D": 6.3889199653375215, "J_D": 155.41709997367855, "W_D_1KI": 0.6841850466199959, "J_D_1KI": 0.0732689062561572}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_040.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_040.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_040", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89658, "MATRIX_DENSITY": 9.105647807962247e-05, "TIME_S": 2.2487361431121826}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89657, 89657, 89658]),
col_indices=tensor([ 106, 329, 1040, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89658, layout=torch.sparse_csr)
tensor([0.1229, 0.7875, 0.3967, ..., 0.2389, 0.2107, 0.7413])
Matrix Type: SuiteSparse
Matrix: as-caida_G_040
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89658
Density: 9.105647807962247e-05
Time: 2.2487361431121826 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 9338 -m matrices/as-caida_pruned/as-caida_G_040.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_040", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89658, "MATRIX_DENSITY": 9.105647807962247e-05, "TIME_S": 20.749686002731323}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89657, 89657, 89658]),
col_indices=tensor([ 106, 329, 1040, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89658, layout=torch.sparse_csr)
tensor([0.4417, 0.5816, 0.3830, ..., 0.9748, 0.5625, 0.5118])
Matrix Type: SuiteSparse
Matrix: as-caida_G_040
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89658
Density: 9.105647807962247e-05
Time: 20.749686002731323 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89657, 89657, 89658]),
col_indices=tensor([ 106, 329, 1040, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89658, layout=torch.sparse_csr)
tensor([0.4417, 0.5816, 0.3830, ..., 0.9748, 0.5625, 0.5118])
Matrix Type: SuiteSparse
Matrix: as-caida_G_040
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89658
Density: 9.105647807962247e-05
Time: 20.749686002731323 seconds
[20.28, 20.52, 20.44, 20.32, 20.24, 20.44, 20.36, 20.56, 20.72, 20.76]
[20.6, 20.44, 20.44, 21.56, 23.72, 30.96, 32.52, 33.28, 31.88, 30.6, 25.08, 24.84, 25.16, 25.08, 25.08, 25.24, 25.2, 25.2, 25.32, 25.72, 25.56, 25.36, 25.12, 24.64]
24.32603645324707
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9338, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_040', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 89658, 'MATRIX_DENSITY': 9.105647807962247e-05, 'TIME_S': 20.749686002731323, 'TIME_S_1KI': 2.222069608345612, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 604.0135382080078, 'W': 24.829919965337524}
[20.28, 20.52, 20.44, 20.32, 20.24, 20.44, 20.36, 20.56, 20.72, 20.76, 20.72, 20.52, 20.76, 20.6, 20.6, 20.32, 20.4, 20.4, 20.44, 20.6]
368.82000000000005
18.441000000000003
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9338, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_040', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 89658, 'MATRIX_DENSITY': 9.105647807962247e-05, 'TIME_S': 20.749686002731323, 'TIME_S_1KI': 2.222069608345612, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 604.0135382080078, 'W': 24.829919965337524, 'J_1KI': 64.68339453930263, 'W_1KI': 2.659019058185642, 'W_D': 6.3889199653375215, 'J_D': 155.41709997367855, 'W_D_1KI': 0.6841850466199959, 'J_D_1KI': 0.0732689062561572}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_045.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 9478, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_045", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89152, "MATRIX_DENSITY": 9.054258553341032e-05, "TIME_S": 20.637967348098755, "TIME_S_1KI": 2.1774601548954164, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 628.3440246582034, "W": 24.768273200628375, "J_1KI": 66.29500154655025, "W_1KI": 2.6132383625900375, "W_D": 6.418273200628374, "J_D": 162.8245772957804, "W_D_1KI": 0.6771759021553464, "J_D_1KI": 0.07144713042364913}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_045.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_045.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_045", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89152, "MATRIX_DENSITY": 9.054258553341032e-05, "TIME_S": 2.21557879447937}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89150, 89150, 89152]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89152, layout=torch.sparse_csr)
tensor([0.8853, 0.7709, 0.2706, ..., 0.2431, 0.7108, 0.8769])
Matrix Type: SuiteSparse
Matrix: as-caida_G_045
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89152
Density: 9.054258553341032e-05
Time: 2.21557879447937 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 9478 -m matrices/as-caida_pruned/as-caida_G_045.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_045", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89152, "MATRIX_DENSITY": 9.054258553341032e-05, "TIME_S": 20.637967348098755}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89150, 89150, 89152]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89152, layout=torch.sparse_csr)
tensor([0.1963, 0.0254, 0.9151, ..., 0.5981, 0.5083, 0.3851])
Matrix Type: SuiteSparse
Matrix: as-caida_G_045
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89152
Density: 9.054258553341032e-05
Time: 20.637967348098755 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89150, 89150, 89152]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89152, layout=torch.sparse_csr)
tensor([0.1963, 0.0254, 0.9151, ..., 0.5981, 0.5083, 0.3851])
Matrix Type: SuiteSparse
Matrix: as-caida_G_045
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89152
Density: 9.054258553341032e-05
Time: 20.637967348098755 seconds
[20.8, 20.56, 20.24, 20.28, 20.4, 20.36, 20.48, 20.72, 20.44, 20.2]
[20.2, 20.2, 20.2, 21.48, 23.16, 29.44, 30.2, 30.84, 30.12, 30.0, 25.2, 25.64, 25.68, 25.68, 25.8, 26.16, 26.36, 26.04, 25.88, 25.72, 25.44, 25.44, 25.72, 25.6, 25.6]
25.36890721321106
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9478, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_045', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 89152, 'MATRIX_DENSITY': 9.054258553341032e-05, 'TIME_S': 20.637967348098755, 'TIME_S_1KI': 2.1774601548954164, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 628.3440246582034, 'W': 24.768273200628375}
[20.8, 20.56, 20.24, 20.28, 20.4, 20.36, 20.48, 20.72, 20.44, 20.2, 20.6, 20.44, 20.4, 20.4, 20.2, 20.32, 20.24, 20.2, 20.36, 20.32]
367.0
18.35
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9478, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_045', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 89152, 'MATRIX_DENSITY': 9.054258553341032e-05, 'TIME_S': 20.637967348098755, 'TIME_S_1KI': 2.1774601548954164, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 628.3440246582034, 'W': 24.768273200628375, 'J_1KI': 66.29500154655025, 'W_1KI': 2.6132383625900375, 'W_D': 6.418273200628374, 'J_D': 162.8245772957804, 'W_D_1KI': 0.6771759021553464, 'J_D_1KI': 0.07144713042364913}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_050.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 9122, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_050", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 90392, "MATRIX_DENSITY": 9.180192695100532e-05, "TIME_S": 20.173283338546753, "TIME_S_1KI": 2.2114978446115714, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 598.2593908691406, "W": 24.587966612408714, "J_1KI": 65.58423491220572, "W_1KI": 2.6954578614787015, "W_D": 6.4099666124087165, "J_D": 155.96339386177067, "W_D_1KI": 0.7026931169051432, "J_D_1KI": 0.07703279071531936}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_050.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_050.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_050", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 90392, "MATRIX_DENSITY": 9.180192695100532e-05, "TIME_S": 2.3019206523895264}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 90390, 90390, 90392]),
col_indices=tensor([ 5326, 106, 329, ..., 882, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=90392, layout=torch.sparse_csr)
tensor([0.4612, 0.9596, 0.7932, ..., 0.5334, 0.6140, 0.6916])
Matrix Type: SuiteSparse
Matrix: as-caida_G_050
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 90392
Density: 9.180192695100532e-05
Time: 2.3019206523895264 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 9122 -m matrices/as-caida_pruned/as-caida_G_050.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_050", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 90392, "MATRIX_DENSITY": 9.180192695100532e-05, "TIME_S": 20.173283338546753}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 90390, 90390, 90392]),
col_indices=tensor([ 5326, 106, 329, ..., 882, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=90392, layout=torch.sparse_csr)
tensor([0.9154, 0.6769, 0.9760, ..., 0.3153, 0.3040, 0.9723])
Matrix Type: SuiteSparse
Matrix: as-caida_G_050
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 90392
Density: 9.180192695100532e-05
Time: 20.173283338546753 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 90390, 90390, 90392]),
col_indices=tensor([ 5326, 106, 329, ..., 882, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=90392, layout=torch.sparse_csr)
tensor([0.9154, 0.6769, 0.9760, ..., 0.3153, 0.3040, 0.9723])
Matrix Type: SuiteSparse
Matrix: as-caida_G_050
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 90392
Density: 9.180192695100532e-05
Time: 20.173283338546753 seconds
[19.8, 19.84, 19.88, 19.88, 20.08, 20.16, 20.24, 20.6, 20.48, 20.8]
[21.12, 21.08, 24.36, 26.16, 27.76, 31.28, 31.28, 32.04, 29.04, 28.04, 24.44, 24.44, 24.52, 24.4, 24.48, 24.08, 24.12, 24.2, 24.24, 24.24, 24.28, 24.52, 24.48, 24.32]
24.33138942718506
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9122, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_050', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 90392, 'MATRIX_DENSITY': 9.180192695100532e-05, 'TIME_S': 20.173283338546753, 'TIME_S_1KI': 2.2114978446115714, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 598.2593908691406, 'W': 24.587966612408714}
[19.8, 19.84, 19.88, 19.88, 20.08, 20.16, 20.24, 20.6, 20.48, 20.8, 20.28, 20.24, 20.4, 20.4, 20.36, 20.28, 20.12, 19.92, 20.16, 20.16]
363.55999999999995
18.177999999999997
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9122, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_050', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 90392, 'MATRIX_DENSITY': 9.180192695100532e-05, 'TIME_S': 20.173283338546753, 'TIME_S_1KI': 2.2114978446115714, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 598.2593908691406, 'W': 24.587966612408714, 'J_1KI': 65.58423491220572, 'W_1KI': 2.6954578614787015, 'W_D': 6.4099666124087165, 'J_D': 155.96339386177067, 'W_D_1KI': 0.7026931169051432, 'J_D_1KI': 0.07703279071531936}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_055.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 9445, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_055", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 91476, "MATRIX_DENSITY": 9.290283509348351e-05, "TIME_S": 21.90527892112732, "TIME_S_1KI": 2.3192460477636128, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 600.4606732463836, "W": 24.72266919997052, "J_1KI": 63.574449258484236, "W_1KI": 2.6175404129137663, "W_D": 6.193669199970518, "J_D": 150.43095660901056, "W_D_1KI": 0.6557616940148775, "J_D_1KI": 0.06942950704233748}

85
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_055.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_055.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_055", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 91476, "MATRIX_DENSITY": 9.290283509348351e-05, "TIME_S": 2.3901093006134033}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 91475, 91475, 91476]),
col_indices=tensor([21783, 106, 329, ..., 160, 882, 17255]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=91476, layout=torch.sparse_csr)
tensor([0.5831, 0.1214, 0.4741, ..., 0.1774, 0.8133, 0.4631])
Matrix Type: SuiteSparse
Matrix: as-caida_G_055
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 91476
Density: 9.290283509348351e-05
Time: 2.3901093006134033 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8786 -m matrices/as-caida_pruned/as-caida_G_055.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_055", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 91476, "MATRIX_DENSITY": 9.290283509348351e-05, "TIME_S": 19.53348445892334}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 91475, 91475, 91476]),
col_indices=tensor([21783, 106, 329, ..., 160, 882, 17255]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=91476, layout=torch.sparse_csr)
tensor([0.2333, 0.7547, 0.3714, ..., 0.2114, 0.9053, 0.3045])
Matrix Type: SuiteSparse
Matrix: as-caida_G_055
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 91476
Density: 9.290283509348351e-05
Time: 19.53348445892334 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 9445 -m matrices/as-caida_pruned/as-caida_G_055.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_055", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 91476, "MATRIX_DENSITY": 9.290283509348351e-05, "TIME_S": 21.90527892112732}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 91475, 91475, 91476]),
col_indices=tensor([21783, 106, 329, ..., 160, 882, 17255]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=91476, layout=torch.sparse_csr)
tensor([0.7742, 0.0253, 0.0492, ..., 0.9424, 0.6447, 0.9766])
Matrix Type: SuiteSparse
Matrix: as-caida_G_055
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 91476
Density: 9.290283509348351e-05
Time: 21.90527892112732 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 91475, 91475, 91476]),
col_indices=tensor([21783, 106, 329, ..., 160, 882, 17255]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=91476, layout=torch.sparse_csr)
tensor([0.7742, 0.0253, 0.0492, ..., 0.9424, 0.6447, 0.9766])
Matrix Type: SuiteSparse
Matrix: as-caida_G_055
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 91476
Density: 9.290283509348351e-05
Time: 21.90527892112732 seconds
[20.28, 20.24, 20.28, 20.56, 20.72, 20.8, 20.8, 20.96, 20.96, 20.76]
[20.6, 20.56, 20.6, 22.4, 23.28, 29.48, 30.4, 30.72, 29.88, 29.88, 25.44, 25.32, 25.32, 25.6, 25.48, 25.2, 25.44, 25.72, 25.52, 25.8, 26.04, 26.04, 25.84, 25.64]
24.2878577709198
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9445, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_055', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 91476, 'MATRIX_DENSITY': 9.290283509348351e-05, 'TIME_S': 21.90527892112732, 'TIME_S_1KI': 2.3192460477636128, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 600.4606732463836, 'W': 24.72266919997052}
[20.28, 20.24, 20.28, 20.56, 20.72, 20.8, 20.8, 20.96, 20.96, 20.76, 20.32, 20.56, 20.8, 20.72, 20.64, 20.52, 20.44, 20.36, 20.28, 20.52]
370.58000000000004
18.529000000000003
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9445, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_055', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 91476, 'MATRIX_DENSITY': 9.290283509348351e-05, 'TIME_S': 21.90527892112732, 'TIME_S_1KI': 2.3192460477636128, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 600.4606732463836, 'W': 24.72266919997052, 'J_1KI': 63.574449258484236, 'W_1KI': 2.6175404129137663, 'W_D': 6.193669199970518, 'J_D': 150.43095660901056, 'W_D_1KI': 0.6557616940148775, 'J_D_1KI': 0.06942950704233748}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_060.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 9065, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_060", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 94180, "MATRIX_DENSITY": 9.564901186217454e-05, "TIME_S": 21.8990478515625, "TIME_S_1KI": 2.4157802373483177, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 595.6967398834229, "W": 24.49063819578549, "J_1KI": 65.71392607649453, "W_1KI": 2.7016699609250403, "W_D": 6.139638195785491, "J_D": 149.33716418719297, "W_D_1KI": 0.6772904794027017, "J_D_1KI": 0.07471489017128535}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_060.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_060.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_060", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 94180, "MATRIX_DENSITY": 9.564901186217454e-05, "TIME_S": 2.316600799560547}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 94180, 94180, 94180]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=94180, layout=torch.sparse_csr)
tensor([0.0765, 0.2337, 0.8130, ..., 0.9153, 0.0336, 0.4203])
Matrix Type: SuiteSparse
Matrix: as-caida_G_060
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 94180
Density: 9.564901186217454e-05
Time: 2.316600799560547 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 9065 -m matrices/as-caida_pruned/as-caida_G_060.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_060", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 94180, "MATRIX_DENSITY": 9.564901186217454e-05, "TIME_S": 21.8990478515625}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 94180, 94180, 94180]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=94180, layout=torch.sparse_csr)
tensor([0.7137, 0.0725, 0.7669, ..., 0.5328, 0.1897, 0.1802])
Matrix Type: SuiteSparse
Matrix: as-caida_G_060
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 94180
Density: 9.564901186217454e-05
Time: 21.8990478515625 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 94180, 94180, 94180]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=94180, layout=torch.sparse_csr)
tensor([0.7137, 0.0725, 0.7669, ..., 0.5328, 0.1897, 0.1802])
Matrix Type: SuiteSparse
Matrix: as-caida_G_060
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 94180
Density: 9.564901186217454e-05
Time: 21.8990478515625 seconds
[20.2, 20.2, 20.04, 20.24, 20.08, 20.2, 20.48, 20.88, 20.64, 20.68]
[20.56, 19.96, 20.72, 23.48, 23.48, 30.04, 31.32, 32.08, 30.72, 30.36, 24.96, 24.96, 25.16, 25.0, 24.68, 24.52, 24.52, 24.52, 24.56, 24.56, 24.84, 24.88, 25.28, 25.28]
24.32344698905945
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9065, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_060', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 94180, 'MATRIX_DENSITY': 9.564901186217454e-05, 'TIME_S': 21.8990478515625, 'TIME_S_1KI': 2.4157802373483177, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 595.6967398834229, 'W': 24.49063819578549}
[20.2, 20.2, 20.04, 20.24, 20.08, 20.2, 20.48, 20.88, 20.64, 20.68, 20.24, 20.28, 20.72, 20.92, 20.6, 20.52, 20.52, 20.4, 19.84, 19.8]
367.02
18.351
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9065, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_060', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 94180, 'MATRIX_DENSITY': 9.564901186217454e-05, 'TIME_S': 21.8990478515625, 'TIME_S_1KI': 2.4157802373483177, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 595.6967398834229, 'W': 24.49063819578549, 'J_1KI': 65.71392607649453, 'W_1KI': 2.7016699609250403, 'W_D': 6.139638195785491, 'J_D': 149.33716418719297, 'W_D_1KI': 0.6772904794027017, 'J_D_1KI': 0.07471489017128535}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_065.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 9003, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_065", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 95068, "MATRIX_DENSITY": 9.655086281283934e-05, "TIME_S": 20.59884262084961, "TIME_S_1KI": 2.287997625330402, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 595.7279040813446, "W": 24.476738666267384, "J_1KI": 66.16993269813891, "W_1KI": 2.7187313857900013, "W_D": 6.194738666267384, "J_D": 150.77084951162337, "W_D_1KI": 0.6880749379392851, "J_D_1KI": 0.07642729511710375}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_065.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_065.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_065", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 95068, "MATRIX_DENSITY": 9.655086281283934e-05, "TIME_S": 2.3325326442718506}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 95068, 95068, 95068]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=95068, layout=torch.sparse_csr)
tensor([0.6782, 0.7733, 0.7625, ..., 0.9659, 0.3257, 0.6368])
Matrix Type: SuiteSparse
Matrix: as-caida_G_065
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 95068
Density: 9.655086281283934e-05
Time: 2.3325326442718506 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 9003 -m matrices/as-caida_pruned/as-caida_G_065.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_065", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 95068, "MATRIX_DENSITY": 9.655086281283934e-05, "TIME_S": 20.59884262084961}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 95068, 95068, 95068]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=95068, layout=torch.sparse_csr)
tensor([0.8368, 0.9690, 0.7119, ..., 0.9803, 0.0517, 0.2837])
Matrix Type: SuiteSparse
Matrix: as-caida_G_065
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 95068
Density: 9.655086281283934e-05
Time: 20.59884262084961 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 95068, 95068, 95068]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=95068, layout=torch.sparse_csr)
tensor([0.8368, 0.9690, 0.7119, ..., 0.9803, 0.0517, 0.2837])
Matrix Type: SuiteSparse
Matrix: as-caida_G_065
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 95068
Density: 9.655086281283934e-05
Time: 20.59884262084961 seconds
[20.6, 20.4, 20.32, 20.12, 20.24, 20.48, 20.8, 20.92, 20.96, 20.84]
[20.84, 20.68, 20.28, 23.6, 26.16, 31.6, 32.72, 33.6, 29.72, 27.6, 24.84, 25.0, 24.84, 24.84, 24.72, 24.64, 24.2, 24.08, 24.2, 24.08, 24.2, 24.52, 24.56, 24.6]
24.33853268623352
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9003, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_065', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 95068, 'MATRIX_DENSITY': 9.655086281283934e-05, 'TIME_S': 20.59884262084961, 'TIME_S_1KI': 2.287997625330402, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 595.7279040813446, 'W': 24.476738666267384}
[20.6, 20.4, 20.32, 20.12, 20.24, 20.48, 20.8, 20.92, 20.96, 20.84, 20.2, 20.04, 20.0, 20.0, 19.92, 19.84, 20.12, 20.2, 20.36, 20.2]
365.64
18.282
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 9003, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_065', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 95068, 'MATRIX_DENSITY': 9.655086281283934e-05, 'TIME_S': 20.59884262084961, 'TIME_S_1KI': 2.287997625330402, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 595.7279040813446, 'W': 24.476738666267384, 'J_1KI': 66.16993269813891, 'W_1KI': 2.7187313857900013, 'W_D': 6.194738666267384, 'J_D': 150.77084951162337, 'W_D_1KI': 0.6880749379392851, 'J_D_1KI': 0.07642729511710375}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_070.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 10632, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_070", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 78684, "MATRIX_DENSITY": 7.991130653390679e-05, "TIME_S": 21.13743305206299, "TIME_S_1KI": 1.9880956595243593, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 577.3553030776978, "W": 23.727390421920692, "J_1KI": 54.30354618864728, "W_1KI": 2.231695863611803, "W_D": 5.278390421920694, "J_D": 128.438342675686, "W_D_1KI": 0.4964626055230149, "J_D_1KI": 0.04669512843519704}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_070.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_070.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_070", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 78684, "MATRIX_DENSITY": 7.991130653390679e-05, "TIME_S": 1.9750430583953857}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 78684, 78684, 78684]),
col_indices=tensor([ 106, 329, 1040, ..., 16263, 2242, 2242]),
values=tensor([1., 1., 1., ..., 3., 1., 1.]), size=(31379, 31379),
nnz=78684, layout=torch.sparse_csr)
tensor([0.2425, 0.1579, 0.4433, ..., 0.4904, 0.0503, 0.2838])
Matrix Type: SuiteSparse
Matrix: as-caida_G_070
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 78684
Density: 7.991130653390679e-05
Time: 1.9750430583953857 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 10632 -m matrices/as-caida_pruned/as-caida_G_070.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_070", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 78684, "MATRIX_DENSITY": 7.991130653390679e-05, "TIME_S": 21.13743305206299}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 78684, 78684, 78684]),
col_indices=tensor([ 106, 329, 1040, ..., 16263, 2242, 2242]),
values=tensor([1., 1., 1., ..., 3., 1., 1.]), size=(31379, 31379),
nnz=78684, layout=torch.sparse_csr)
tensor([0.2555, 0.5888, 0.5922, ..., 0.9980, 0.6558, 0.7472])
Matrix Type: SuiteSparse
Matrix: as-caida_G_070
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 78684
Density: 7.991130653390679e-05
Time: 21.13743305206299 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 78684, 78684, 78684]),
col_indices=tensor([ 106, 329, 1040, ..., 16263, 2242, 2242]),
values=tensor([1., 1., 1., ..., 3., 1., 1.]), size=(31379, 31379),
nnz=78684, layout=torch.sparse_csr)
tensor([0.2555, 0.5888, 0.5922, ..., 0.9980, 0.6558, 0.7472])
Matrix Type: SuiteSparse
Matrix: as-caida_G_070
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 78684
Density: 7.991130653390679e-05
Time: 21.13743305206299 seconds
[20.56, 20.44, 20.48, 20.48, 20.28, 20.28, 20.36, 20.28, 20.6, 20.6]
[20.6, 20.68, 20.76, 23.92, 25.08, 28.64, 29.68, 29.96, 27.36, 27.36, 24.44, 24.24, 24.44, 24.44, 24.2, 24.16, 24.12, 24.08, 23.8, 23.8, 23.68, 24.04, 24.12, 24.08]
24.33286142349243
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 10632, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_070', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 78684, 'MATRIX_DENSITY': 7.991130653390679e-05, 'TIME_S': 21.13743305206299, 'TIME_S_1KI': 1.9880956595243593, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 577.3553030776978, 'W': 23.727390421920692}
[20.56, 20.44, 20.48, 20.48, 20.28, 20.28, 20.36, 20.28, 20.6, 20.6, 20.08, 20.2, 20.64, 20.64, 20.88, 20.72, 20.56, 20.64, 20.64, 20.48]
368.97999999999996
18.448999999999998
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 10632, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_070', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 78684, 'MATRIX_DENSITY': 7.991130653390679e-05, 'TIME_S': 21.13743305206299, 'TIME_S_1KI': 1.9880956595243593, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 577.3553030776978, 'W': 23.727390421920692, 'J_1KI': 54.30354618864728, 'W_1KI': 2.231695863611803, 'W_D': 5.278390421920694, 'J_D': 128.438342675686, 'W_D_1KI': 0.4964626055230149, 'J_D_1KI': 0.04669512843519704}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_075.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8584, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_075", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 97492, "MATRIX_DENSITY": 9.901267216465406e-05, "TIME_S": 20.15437150001526, "TIME_S_1KI": 2.3478997553605847, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 602.6740414428712, "W": 23.734224847984613, "J_1KI": 70.20899830415554, "W_1KI": 2.764937657034554, "W_D": 5.527224847984613, "J_D": 140.35069434261328, "W_D_1KI": 0.6438985144436874, "J_D_1KI": 0.07501147651953488}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_075.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_075.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_075", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 97492, "MATRIX_DENSITY": 9.901267216465406e-05, "TIME_S": 2.446396589279175}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 97491, 97491, 97492]),
col_indices=tensor([22754, 22754, 106, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=97492, layout=torch.sparse_csr)
tensor([0.6358, 0.8814, 0.7109, ..., 0.5208, 0.0272, 0.7061])
Matrix Type: SuiteSparse
Matrix: as-caida_G_075
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 97492
Density: 9.901267216465406e-05
Time: 2.446396589279175 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8584 -m matrices/as-caida_pruned/as-caida_G_075.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_075", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 97492, "MATRIX_DENSITY": 9.901267216465406e-05, "TIME_S": 20.15437150001526}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 97491, 97491, 97492]),
col_indices=tensor([22754, 22754, 106, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=97492, layout=torch.sparse_csr)
tensor([0.5045, 0.1890, 0.1651, ..., 0.8675, 0.9058, 0.0413])
Matrix Type: SuiteSparse
Matrix: as-caida_G_075
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 97492
Density: 9.901267216465406e-05
Time: 20.15437150001526 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 97491, 97491, 97492]),
col_indices=tensor([22754, 22754, 106, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=97492, layout=torch.sparse_csr)
tensor([0.5045, 0.1890, 0.1651, ..., 0.8675, 0.9058, 0.0413])
Matrix Type: SuiteSparse
Matrix: as-caida_G_075
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 97492
Density: 9.901267216465406e-05
Time: 20.15437150001526 seconds
[20.16, 20.08, 19.92, 20.24, 20.4, 20.36, 20.44, 20.68, 20.6, 20.36]
[20.48, 20.56, 21.16, 22.32, 22.32, 25.96, 27.12, 27.92, 27.96, 27.6, 24.84, 24.64, 24.44, 24.44, 24.76, 24.92, 24.92, 24.8, 24.8, 24.96, 25.0, 24.88, 25.04, 25.08, 24.64]
25.39261531829834
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8584, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_075', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 97492, 'MATRIX_DENSITY': 9.901267216465406e-05, 'TIME_S': 20.15437150001526, 'TIME_S_1KI': 2.3478997553605847, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 602.6740414428712, 'W': 23.734224847984613}
[20.16, 20.08, 19.92, 20.24, 20.4, 20.36, 20.44, 20.68, 20.6, 20.36, 20.04, 20.0, 20.0, 20.04, 20.12, 20.12, 20.36, 20.28, 20.2, 20.04]
364.14
18.207
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8584, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_075', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 97492, 'MATRIX_DENSITY': 9.901267216465406e-05, 'TIME_S': 20.15437150001526, 'TIME_S_1KI': 2.3478997553605847, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 602.6740414428712, 'W': 23.734224847984613, 'J_1KI': 70.20899830415554, 'W_1KI': 2.764937657034554, 'W_D': 5.527224847984613, 'J_D': 140.35069434261328, 'W_D_1KI': 0.6438985144436874, 'J_D_1KI': 0.07501147651953488}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_080.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8794, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_080", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 98112, "MATRIX_DENSITY": 9.964234287345156e-05, "TIME_S": 23.002132892608643, "TIME_S_1KI": 2.615662143803575, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 576.0093463134767, "W": 23.661706671572514, "J_1KI": 65.5002668084463, "W_1KI": 2.6906648478021964, "W_D": 5.470706671572515, "J_D": 133.17628425979632, "W_D_1KI": 0.6220953686118393, "J_D_1KI": 0.07074088794767333}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_080.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_080.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_080", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 98112, "MATRIX_DENSITY": 9.964234287345156e-05, "TIME_S": 2.387787103652954}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 98111, 98111, 98112]),
col_indices=tensor([22754, 22754, 106, ..., 4133, 31329, 12170]),
values=tensor([1., 1., 1., ..., 3., 3., 1.]), size=(31379, 31379),
nnz=98112, layout=torch.sparse_csr)
tensor([0.1559, 0.4913, 0.6290, ..., 0.0122, 0.8386, 0.3778])
Matrix Type: SuiteSparse
Matrix: as-caida_G_080
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 98112
Density: 9.964234287345156e-05
Time: 2.387787103652954 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8794 -m matrices/as-caida_pruned/as-caida_G_080.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_080", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 98112, "MATRIX_DENSITY": 9.964234287345156e-05, "TIME_S": 23.002132892608643}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 98111, 98111, 98112]),
col_indices=tensor([22754, 22754, 106, ..., 4133, 31329, 12170]),
values=tensor([1., 1., 1., ..., 3., 3., 1.]), size=(31379, 31379),
nnz=98112, layout=torch.sparse_csr)
tensor([0.4057, 0.4897, 0.2712, ..., 0.1739, 0.4440, 0.2669])
Matrix Type: SuiteSparse
Matrix: as-caida_G_080
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 98112
Density: 9.964234287345156e-05
Time: 23.002132892608643 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 98111, 98111, 98112]),
col_indices=tensor([22754, 22754, 106, ..., 4133, 31329, 12170]),
values=tensor([1., 1., 1., ..., 3., 3., 1.]), size=(31379, 31379),
nnz=98112, layout=torch.sparse_csr)
tensor([0.4057, 0.4897, 0.2712, ..., 0.1739, 0.4440, 0.2669])
Matrix Type: SuiteSparse
Matrix: as-caida_G_080
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 98112
Density: 9.964234287345156e-05
Time: 23.002132892608643 seconds
[20.68, 20.36, 20.16, 20.2, 20.08, 20.08, 20.2, 20.2, 20.24, 20.36]
[20.36, 20.48, 20.56, 21.56, 22.68, 25.84, 27.12, 27.52, 27.12, 24.92, 24.92, 25.04, 24.88, 24.84, 25.12, 25.16, 25.16, 25.24, 25.52, 25.48, 25.32, 25.08, 25.08, 25.12]
24.343524932861328
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8794, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_080', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 98112, 'MATRIX_DENSITY': 9.964234287345156e-05, 'TIME_S': 23.002132892608643, 'TIME_S_1KI': 2.615662143803575, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 576.0093463134767, 'W': 23.661706671572514}
[20.68, 20.36, 20.16, 20.2, 20.08, 20.08, 20.2, 20.2, 20.24, 20.36, 20.16, 20.2, 20.28, 20.28, 20.24, 20.32, 20.12, 20.04, 20.16, 20.12]
363.82
18.191
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8794, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_080', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 98112, 'MATRIX_DENSITY': 9.964234287345156e-05, 'TIME_S': 23.002132892608643, 'TIME_S_1KI': 2.615662143803575, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 576.0093463134767, 'W': 23.661706671572514, 'J_1KI': 65.5002668084463, 'W_1KI': 2.6906648478021964, 'W_D': 5.470706671572515, 'J_D': 133.17628425979632, 'W_D_1KI': 0.6220953686118393, 'J_D_1KI': 0.07074088794767333}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_085.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8691, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_085", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 99166, "MATRIX_DENSITY": 0.0001007127830784073, "TIME_S": 20.764369010925293, "TIME_S_1KI": 2.3891806479030366, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 571.7078634071349, "W": 23.505529827290687, "J_1KI": 65.78159744645437, "W_1KI": 2.704582881980288, "W_D": 5.296529827290687, "J_D": 128.823633131504, "W_D_1KI": 0.6094269735692885, "J_D_1KI": 0.07012161702557686}

65
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_085.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_085.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_085", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 99166, "MATRIX_DENSITY": 0.0001007127830784073, "TIME_S": 2.4162235260009766}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 99165, 99165, 99166]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=99166, layout=torch.sparse_csr)
tensor([0.9352, 0.7507, 0.1402, ..., 0.1494, 0.4151, 0.0213])
Matrix Type: SuiteSparse
Matrix: as-caida_G_085
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 99166
Density: 0.0001007127830784073
Time: 2.4162235260009766 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8691 -m matrices/as-caida_pruned/as-caida_G_085.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_085", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 99166, "MATRIX_DENSITY": 0.0001007127830784073, "TIME_S": 20.764369010925293}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 99165, 99165, 99166]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=99166, layout=torch.sparse_csr)
tensor([0.9390, 0.2795, 0.0601, ..., 0.8952, 0.5532, 0.1504])
Matrix Type: SuiteSparse
Matrix: as-caida_G_085
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 99166
Density: 0.0001007127830784073
Time: 20.764369010925293 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 99165, 99165, 99166]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=99166, layout=torch.sparse_csr)
tensor([0.9390, 0.2795, 0.0601, ..., 0.8952, 0.5532, 0.1504])
Matrix Type: SuiteSparse
Matrix: as-caida_G_085
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 99166
Density: 0.0001007127830784073
Time: 20.764369010925293 seconds
[20.4, 20.4, 20.24, 20.32, 20.4, 20.36, 20.2, 20.12, 20.08, 20.08]
[20.08, 20.12, 20.12, 22.72, 23.68, 27.52, 28.64, 27.72, 27.68, 24.8, 24.72, 24.36, 24.32, 24.32, 24.28, 24.24, 24.24, 24.6, 24.52, 24.8, 24.76, 24.56, 24.72, 24.6]
24.32227087020874
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8691, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_085', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 99166, 'MATRIX_DENSITY': 0.0001007127830784073, 'TIME_S': 20.764369010925293, 'TIME_S_1KI': 2.3891806479030366, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 571.7078634071349, 'W': 23.505529827290687}
[20.4, 20.4, 20.24, 20.32, 20.4, 20.36, 20.2, 20.12, 20.08, 20.08, 20.12, 20.08, 20.08, 20.24, 20.4, 20.48, 20.24, 20.12, 20.08, 20.08]
364.18
18.209
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8691, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_085', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 99166, 'MATRIX_DENSITY': 0.0001007127830784073, 'TIME_S': 20.764369010925293, 'TIME_S_1KI': 2.3891806479030366, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 571.7078634071349, 'W': 23.505529827290687, 'J_1KI': 65.78159744645437, 'W_1KI': 2.704582881980288, 'W_D': 5.296529827290687, 'J_D': 128.823633131504, 'W_D_1KI': 0.6094269735692885, 'J_D_1KI': 0.07012161702557686}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_090.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8443, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_090", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 100924, "MATRIX_DENSITY": 0.00010249820421722343, "TIME_S": 20.46173071861267, "TIME_S_1KI": 2.423514238850251, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 617.1725379943847, "W": 25.383726169427987, "J_1KI": 73.09872533393163, "W_1KI": 3.0064818393258306, "W_D": 7.037726169427991, "J_D": 171.11322792816162, "W_D_1KI": 0.8335575233244097, "J_D_1KI": 0.09872764696487145}

68
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_090.output Normal file
View File

@ -0,0 +1,68 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_090.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_090", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 100924, "MATRIX_DENSITY": 0.00010249820421722343, "TIME_S": 2.4871225357055664}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 100923, 100923,
100924]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=100924, layout=torch.sparse_csr)
tensor([0.3922, 0.8557, 0.2310, ..., 0.9809, 0.9954, 0.7931])
Matrix Type: SuiteSparse
Matrix: as-caida_G_090
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 100924
Density: 0.00010249820421722343
Time: 2.4871225357055664 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8443 -m matrices/as-caida_pruned/as-caida_G_090.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_090", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 100924, "MATRIX_DENSITY": 0.00010249820421722343, "TIME_S": 20.46173071861267}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 100923, 100923,
100924]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=100924, layout=torch.sparse_csr)
tensor([0.4649, 0.8948, 0.1887, ..., 0.0336, 0.7739, 0.8142])
Matrix Type: SuiteSparse
Matrix: as-caida_G_090
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 100924
Density: 0.00010249820421722343
Time: 20.46173071861267 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 100923, 100923,
100924]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=100924, layout=torch.sparse_csr)
tensor([0.4649, 0.8948, 0.1887, ..., 0.0336, 0.7739, 0.8142])
Matrix Type: SuiteSparse
Matrix: as-caida_G_090
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 100924
Density: 0.00010249820421722343
Time: 20.46173071861267 seconds
[20.36, 20.16, 20.56, 20.64, 20.64, 20.48, 20.44, 20.52, 20.4, 20.52]
[20.56, 20.24, 20.68, 25.32, 29.76, 33.36, 34.24, 34.24, 31.84, 30.84, 24.56, 24.52, 24.64, 25.0, 25.0, 25.08, 25.36, 25.28, 25.4, 25.4, 25.4, 25.48, 25.04, 24.84]
24.313709259033203
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8443, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_090', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 100924, 'MATRIX_DENSITY': 0.00010249820421722343, 'TIME_S': 20.46173071861267, 'TIME_S_1KI': 2.423514238850251, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 617.1725379943847, 'W': 25.383726169427987}
[20.36, 20.16, 20.56, 20.64, 20.64, 20.48, 20.44, 20.52, 20.4, 20.52, 20.2, 20.24, 20.44, 20.56, 20.56, 20.44, 20.2, 20.08, 20.0, 20.04]
366.91999999999996
18.345999999999997
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8443, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_090', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 100924, 'MATRIX_DENSITY': 0.00010249820421722343, 'TIME_S': 20.46173071861267, 'TIME_S_1KI': 2.423514238850251, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 617.1725379943847, 'W': 25.383726169427987, 'J_1KI': 73.09872533393163, 'W_1KI': 3.0064818393258306, 'W_D': 7.037726169427991, 'J_D': 171.11322792816162, 'W_D_1KI': 0.8335575233244097, 'J_D_1KI': 0.09872764696487145}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_095.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8506, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_095", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102290, "MATRIX_DENSITY": 0.00010388551097241275, "TIME_S": 20.99284601211548, "TIME_S_1KI": 2.4680044688590965, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 586.6492254638672, "W": 24.123598992038087, "J_1KI": 68.96887202725925, "W_1KI": 2.8360685389181852, "W_D": 5.700598992038092, "J_D": 138.62989450550094, "W_D_1KI": 0.670185632734316, "J_D_1KI": 0.07878975226126451}

68
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_095.output Normal file
View File

@ -0,0 +1,68 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_095.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_095", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102290, "MATRIX_DENSITY": 0.00010388551097241275, "TIME_S": 2.4687836170196533}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102289, 102289,
102290]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 25970, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102290, layout=torch.sparse_csr)
tensor([0.5566, 0.3690, 0.1680, ..., 0.0525, 0.8406, 0.9386])
Matrix Type: SuiteSparse
Matrix: as-caida_G_095
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102290
Density: 0.00010388551097241275
Time: 2.4687836170196533 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8506 -m matrices/as-caida_pruned/as-caida_G_095.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_095", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102290, "MATRIX_DENSITY": 0.00010388551097241275, "TIME_S": 20.99284601211548}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102289, 102289,
102290]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 25970, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102290, layout=torch.sparse_csr)
tensor([0.9560, 0.2133, 0.6850, ..., 0.4190, 0.4977, 0.6761])
Matrix Type: SuiteSparse
Matrix: as-caida_G_095
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102290
Density: 0.00010388551097241275
Time: 20.99284601211548 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102289, 102289,
102290]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 25970, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102290, layout=torch.sparse_csr)
tensor([0.9560, 0.2133, 0.6850, ..., 0.4190, 0.4977, 0.6761])
Matrix Type: SuiteSparse
Matrix: as-caida_G_095
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102290
Density: 0.00010388551097241275
Time: 20.99284601211548 seconds
[20.24, 20.32, 20.32, 20.32, 20.28, 20.24, 20.4, 20.32, 20.36, 20.48]
[20.36, 20.24, 21.48, 22.76, 26.32, 27.48, 27.48, 28.48, 27.92, 27.44, 25.08, 25.08, 25.36, 25.36, 25.28, 24.84, 24.92, 24.8, 24.8, 25.24, 25.32, 25.2, 25.16, 24.96]
24.31847858428955
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8506, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_095', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 102290, 'MATRIX_DENSITY': 0.00010388551097241275, 'TIME_S': 20.99284601211548, 'TIME_S_1KI': 2.4680044688590965, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 586.6492254638672, 'W': 24.123598992038087}
[20.24, 20.32, 20.32, 20.32, 20.28, 20.24, 20.4, 20.32, 20.36, 20.48, 20.56, 20.64, 20.76, 20.56, 20.6, 20.76, 20.72, 20.48, 20.48, 20.52]
368.4599999999999
18.422999999999995
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8506, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_095', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 102290, 'MATRIX_DENSITY': 0.00010388551097241275, 'TIME_S': 20.99284601211548, 'TIME_S_1KI': 2.4680044688590965, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 586.6492254638672, 'W': 24.123598992038087, 'J_1KI': 68.96887202725925, 'W_1KI': 2.8360685389181852, 'W_D': 5.700598992038092, 'J_D': 138.62989450550094, 'W_D_1KI': 0.670185632734316, 'J_D_1KI': 0.07878975226126451}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_100.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8467, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_100", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102888, "MATRIX_DENSITY": 0.00010449283852702711, "TIME_S": 20.76715898513794, "TIME_S_1KI": 2.452717489682053, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 624.7324462318422, "W": 24.68048649134715, "J_1KI": 73.78439190171753, "W_1KI": 2.914903329555586, "W_D": 6.425486491347147, "J_D": 162.64711375832567, "W_D_1KI": 0.7588858499287997, "J_D_1KI": 0.0896286583121294}

68
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_100.output Normal file
View File

@ -0,0 +1,68 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_100.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_100", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102888, "MATRIX_DENSITY": 0.00010449283852702711, "TIME_S": 2.4800076484680176}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102886, 102887,
102888]),
col_indices=tensor([ 106, 329, 1040, ..., 25970, 5128, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102888, layout=torch.sparse_csr)
tensor([0.5380, 0.6798, 0.7075, ..., 0.5675, 0.9961, 0.3743])
Matrix Type: SuiteSparse
Matrix: as-caida_G_100
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102888
Density: 0.00010449283852702711
Time: 2.4800076484680176 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8467 -m matrices/as-caida_pruned/as-caida_G_100.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_100", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102888, "MATRIX_DENSITY": 0.00010449283852702711, "TIME_S": 20.76715898513794}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102886, 102887,
102888]),
col_indices=tensor([ 106, 329, 1040, ..., 25970, 5128, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102888, layout=torch.sparse_csr)
tensor([0.6117, 0.2465, 0.6631, ..., 0.0512, 0.3459, 0.0433])
Matrix Type: SuiteSparse
Matrix: as-caida_G_100
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102888
Density: 0.00010449283852702711
Time: 20.76715898513794 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102886, 102887,
102888]),
col_indices=tensor([ 106, 329, 1040, ..., 25970, 5128, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102888, layout=torch.sparse_csr)
tensor([0.6117, 0.2465, 0.6631, ..., 0.0512, 0.3459, 0.0433])
Matrix Type: SuiteSparse
Matrix: as-caida_G_100
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102888
Density: 0.00010449283852702711
Time: 20.76715898513794 seconds
[20.28, 20.28, 20.04, 19.88, 19.64, 19.64, 19.8, 20.08, 20.2, 20.04]
[20.08, 20.08, 23.48, 25.8, 25.8, 32.84, 33.72, 34.6, 30.6, 28.72, 23.88, 23.84, 24.0, 24.24, 24.48, 24.48, 24.48, 24.6, 24.52, 24.24, 24.28, 24.24, 24.16, 24.04, 24.08]
25.312809228897095
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8467, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_100', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 102888, 'MATRIX_DENSITY': 0.00010449283852702711, 'TIME_S': 20.76715898513794, 'TIME_S_1KI': 2.452717489682053, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 624.7324462318422, 'W': 24.68048649134715}
[20.28, 20.28, 20.04, 19.88, 19.64, 19.64, 19.8, 20.08, 20.2, 20.04, 20.72, 20.56, 20.68, 20.6, 20.48, 20.48, 20.52, 20.8, 20.56, 20.68]
365.1
18.255000000000003
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8467, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_100', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 102888, 'MATRIX_DENSITY': 0.00010449283852702711, 'TIME_S': 20.76715898513794, 'TIME_S_1KI': 2.452717489682053, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 624.7324462318422, 'W': 24.68048649134715, 'J_1KI': 73.78439190171753, 'W_1KI': 2.914903329555586, 'W_D': 6.425486491347147, 'J_D': 162.64711375832567, 'W_D_1KI': 0.7588858499287997, 'J_D_1KI': 0.0896286583121294}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_105.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8307, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_105", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104726, "MATRIX_DENSITY": 0.00010635950749923647, "TIME_S": 20.873857975006104, "TIME_S_1KI": 2.5128034157946435, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 597.9551256275176, "W": 23.607426160996845, "J_1KI": 71.98207844318257, "W_1KI": 2.8418714531114535, "W_D": 5.241426160996845, "J_D": 132.76066680002205, "W_D_1KI": 0.6309649886838623, "J_D_1KI": 0.0759558190301989}

68
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_105.output Normal file
View File

@ -0,0 +1,68 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_105.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_105", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104726, "MATRIX_DENSITY": 0.00010635950749923647, "TIME_S": 2.527773141860962}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104725, 104725,
104726]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104726, layout=torch.sparse_csr)
tensor([0.1824, 0.7823, 0.6330, ..., 0.4083, 0.5524, 0.9351])
Matrix Type: SuiteSparse
Matrix: as-caida_G_105
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104726
Density: 0.00010635950749923647
Time: 2.527773141860962 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8307 -m matrices/as-caida_pruned/as-caida_G_105.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_105", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104726, "MATRIX_DENSITY": 0.00010635950749923647, "TIME_S": 20.873857975006104}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104725, 104725,
104726]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104726, layout=torch.sparse_csr)
tensor([0.9834, 0.9632, 0.1557, ..., 0.5772, 0.3745, 0.8872])
Matrix Type: SuiteSparse
Matrix: as-caida_G_105
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104726
Density: 0.00010635950749923647
Time: 20.873857975006104 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104725, 104725,
104726]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104726, layout=torch.sparse_csr)
tensor([0.9834, 0.9632, 0.1557, ..., 0.5772, 0.3745, 0.8872])
Matrix Type: SuiteSparse
Matrix: as-caida_G_105
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104726
Density: 0.00010635950749923647
Time: 20.873857975006104 seconds
[20.96, 20.88, 21.0, 21.72, 20.72, 20.36, 20.32, 20.28, 20.12, 20.32]
[20.4, 20.48, 21.0, 22.72, 23.48, 27.08, 28.0, 28.0, 27.4, 27.12, 24.52, 24.36, 24.36, 24.28, 24.28, 24.24, 24.28, 24.28, 24.32, 24.32, 24.4, 24.6, 24.68, 24.96, 24.84]
25.329111337661743
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8307, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_105', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 104726, 'MATRIX_DENSITY': 0.00010635950749923647, 'TIME_S': 20.873857975006104, 'TIME_S_1KI': 2.5128034157946435, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 597.9551256275176, 'W': 23.607426160996845}
[20.96, 20.88, 21.0, 21.72, 20.72, 20.36, 20.32, 20.28, 20.12, 20.32, 20.4, 20.44, 20.0, 19.88, 19.84, 19.88, 20.16, 20.32, 20.32, 20.48]
367.32
18.366
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8307, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_105', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 104726, 'MATRIX_DENSITY': 0.00010635950749923647, 'TIME_S': 20.873857975006104, 'TIME_S_1KI': 2.5128034157946435, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 597.9551256275176, 'W': 23.607426160996845, 'J_1KI': 71.98207844318257, 'W_1KI': 2.8418714531114535, 'W_D': 5.241426160996845, 'J_D': 132.76066680002205, 'W_D_1KI': 0.6309649886838623, 'J_D_1KI': 0.0759558190301989}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_110.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8349, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_110", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104846, "MATRIX_DENSITY": 0.0001064813792493263, "TIME_S": 21.81509804725647, "TIME_S_1KI": 2.6128995145833596, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 606.0642761802675, "W": 24.910639760816423, "J_1KI": 72.59124160741017, "W_1KI": 2.9836674764422595, "W_D": 6.720639760816418, "J_D": 163.5100387310982, "W_D_1KI": 0.8049634400307124, "J_D_1KI": 0.0964143538185067}

68
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_110.output Normal file
View File

@ -0,0 +1,68 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_110.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_110", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104846, "MATRIX_DENSITY": 0.0001064813792493263, "TIME_S": 2.5151610374450684}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104844, 104844,
104846]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 2616, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104846, layout=torch.sparse_csr)
tensor([0.3789, 0.4285, 0.7113, ..., 0.5357, 0.8339, 0.3006])
Matrix Type: SuiteSparse
Matrix: as-caida_G_110
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104846
Density: 0.0001064813792493263
Time: 2.5151610374450684 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8349 -m matrices/as-caida_pruned/as-caida_G_110.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_110", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104846, "MATRIX_DENSITY": 0.0001064813792493263, "TIME_S": 21.81509804725647}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104844, 104844,
104846]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 2616, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104846, layout=torch.sparse_csr)
tensor([0.7764, 0.6906, 0.1766, ..., 0.9379, 0.5296, 0.5222])
Matrix Type: SuiteSparse
Matrix: as-caida_G_110
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104846
Density: 0.0001064813792493263
Time: 21.81509804725647 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104844, 104844,
104846]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 2616, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104846, layout=torch.sparse_csr)
tensor([0.7764, 0.6906, 0.1766, ..., 0.9379, 0.5296, 0.5222])
Matrix Type: SuiteSparse
Matrix: as-caida_G_110
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104846
Density: 0.0001064813792493263
Time: 21.81509804725647 seconds
[19.96, 19.8, 19.72, 19.72, 20.04, 20.36, 20.32, 20.56, 20.8, 20.8]
[20.56, 20.56, 20.2, 23.44, 25.88, 32.52, 33.72, 34.12, 30.6, 28.04, 25.16, 25.24, 24.88, 24.88, 24.96, 24.92, 25.04, 25.28, 25.08, 25.04, 25.16, 25.2, 24.96, 25.2]
24.329534769058228
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8349, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_110', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 104846, 'MATRIX_DENSITY': 0.0001064813792493263, 'TIME_S': 21.81509804725647, 'TIME_S_1KI': 2.6128995145833596, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 606.0642761802675, 'W': 24.910639760816423}
[19.96, 19.8, 19.72, 19.72, 20.04, 20.36, 20.32, 20.56, 20.8, 20.8, 20.36, 20.2, 20.2, 20.2, 20.36, 20.28, 20.32, 20.24, 20.12, 20.0]
363.80000000000007
18.190000000000005
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8349, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_110', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 104846, 'MATRIX_DENSITY': 0.0001064813792493263, 'TIME_S': 21.81509804725647, 'TIME_S_1KI': 2.6128995145833596, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 606.0642761802675, 'W': 24.910639760816423, 'J_1KI': 72.59124160741017, 'W_1KI': 2.9836674764422595, 'W_D': 6.720639760816418, 'J_D': 163.5100387310982, 'W_D_1KI': 0.8049634400307124, 'J_D_1KI': 0.0964143538185067}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_115.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8049, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_115", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106312, "MATRIX_DENSITY": 0.00010797024579625715, "TIME_S": 20.547222137451172, "TIME_S_1KI": 2.552767068884479, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 603.5182037734984, "W": 24.810850099333518, "J_1KI": 74.98051978798588, "W_1KI": 3.0824760963266886, "W_D": 6.508850099333518, "J_D": 158.32627680444708, "W_D_1KI": 0.8086532611918894, "J_D_1KI": 0.1004663015519803}

68
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_115.output Normal file
View File

@ -0,0 +1,68 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_115.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_115", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106312, "MATRIX_DENSITY": 0.00010797024579625715, "TIME_S": 2.608731508255005}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106311, 106311,
106312]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106312, layout=torch.sparse_csr)
tensor([0.1425, 0.7093, 0.4708, ..., 0.9603, 0.3645, 0.2733])
Matrix Type: SuiteSparse
Matrix: as-caida_G_115
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106312
Density: 0.00010797024579625715
Time: 2.608731508255005 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8049 -m matrices/as-caida_pruned/as-caida_G_115.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_115", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106312, "MATRIX_DENSITY": 0.00010797024579625715, "TIME_S": 20.547222137451172}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106311, 106311,
106312]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106312, layout=torch.sparse_csr)
tensor([0.9041, 0.6122, 0.9574, ..., 0.6613, 0.0965, 0.8368])
Matrix Type: SuiteSparse
Matrix: as-caida_G_115
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106312
Density: 0.00010797024579625715
Time: 20.547222137451172 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106311, 106311,
106312]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106312, layout=torch.sparse_csr)
tensor([0.9041, 0.6122, 0.9574, ..., 0.6613, 0.0965, 0.8368])
Matrix Type: SuiteSparse
Matrix: as-caida_G_115
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106312
Density: 0.00010797024579625715
Time: 20.547222137451172 seconds
[20.2, 20.16, 20.0, 20.24, 20.2, 20.28, 20.32, 20.36, 20.36, 20.4]
[20.4, 20.32, 20.04, 23.64, 25.32, 32.2, 33.56, 33.8, 30.52, 27.4, 24.68, 24.64, 24.64, 24.64, 25.0, 24.92, 24.96, 25.2, 25.28, 25.28, 25.52, 25.56, 25.48, 25.12]
24.324769258499146
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8049, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_115', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 106312, 'MATRIX_DENSITY': 0.00010797024579625715, 'TIME_S': 20.547222137451172, 'TIME_S_1KI': 2.552767068884479, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 603.5182037734984, 'W': 24.810850099333518}
[20.2, 20.16, 20.0, 20.24, 20.2, 20.28, 20.32, 20.36, 20.36, 20.4, 20.68, 20.76, 20.72, 20.72, 20.36, 20.2, 20.16, 20.24, 20.2, 20.24]
366.03999999999996
18.302
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8049, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_115', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 106312, 'MATRIX_DENSITY': 0.00010797024579625715, 'TIME_S': 20.547222137451172, 'TIME_S_1KI': 2.552767068884479, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 603.5182037734984, 'W': 24.810850099333518, 'J_1KI': 74.98051978798588, 'W_1KI': 3.0824760963266886, 'W_D': 6.508850099333518, 'J_D': 158.32627680444708, 'W_D_1KI': 0.8086532611918894, 'J_D_1KI': 0.1004663015519803}

1
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_120.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "CORES": 80, "ITERATIONS": 8105, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_120", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106510, "MATRIX_DENSITY": 0.0001081713341839054, "TIME_S": 20.750195264816284, "TIME_S_1KI": 2.560172148650991, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 630.7465198040009, "W": 24.92526992228312, "J_1KI": 77.8219025051204, "W_1KI": 3.075295487018275, "W_D": 6.416269922283121, "J_D": 162.3669447200298, "W_D_1KI": 0.7916434203927355, "J_D_1KI": 0.09767346334271874}

89
pytorch/output_as-caida_maxcore/altra_max_csr_20_10_10_as-caida_G_120.output Normal file
View File

@ -0,0 +1,89 @@
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 1000 -m matrices/as-caida_pruned/as-caida_G_120.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_120", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106510, "MATRIX_DENSITY": 0.0001081713341839054, "TIME_S": 2.724186420440674}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106509, 106509,
106510]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106510, layout=torch.sparse_csr)
tensor([0.1163, 0.7095, 0.6781, ..., 0.7446, 0.8524, 0.8056])
Matrix Type: SuiteSparse
Matrix: as-caida_G_120
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106510
Density: 0.0001081713341839054
Time: 2.724186420440674 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 7708 -m matrices/as-caida_pruned/as-caida_G_120.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_120", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106510, "MATRIX_DENSITY": 0.0001081713341839054, "TIME_S": 19.970391988754272}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106509, 106509,
106510]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106510, layout=torch.sparse_csr)
tensor([0.8144, 0.1674, 0.8184, ..., 0.8996, 0.3640, 0.9980])
Matrix Type: SuiteSparse
Matrix: as-caida_G_120
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106510
Density: 0.0001081713341839054
Time: 19.970391988754272 seconds
['apptainer', 'run', 'pytorch-altra.sif', '-c', 'numactl --cpunodebind=0 --membind=0 python3 spmv.py suitesparse csr 8105 -m matrices/as-caida_pruned/as-caida_G_120.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_120", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106510, "MATRIX_DENSITY": 0.0001081713341839054, "TIME_S": 20.750195264816284}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106509, 106509,
106510]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106510, layout=torch.sparse_csr)
tensor([0.2321, 0.6812, 0.5513, ..., 0.9888, 0.4472, 0.0437])
Matrix Type: SuiteSparse
Matrix: as-caida_G_120
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106510
Density: 0.0001081713341839054
Time: 20.750195264816284 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106509, 106509,
106510]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106510, layout=torch.sparse_csr)
tensor([0.2321, 0.6812, 0.5513, ..., 0.9888, 0.4472, 0.0437])
Matrix Type: SuiteSparse
Matrix: as-caida_G_120
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106510
Density: 0.0001081713341839054
Time: 20.750195264816284 seconds
[19.92, 20.0, 20.36, 20.36, 20.64, 20.92, 21.12, 21.0, 20.76, 20.36]
[20.44, 20.28, 23.0, 24.16, 31.28, 32.24, 32.24, 33.32, 30.72, 30.24, 24.48, 24.64, 24.6, 24.64, 24.68, 24.68, 24.48, 24.4, 24.4, 24.48, 24.56, 24.28, 24.48, 24.56, 24.44]
25.305504083633423
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8105, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_120', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 106510, 'MATRIX_DENSITY': 0.0001081713341839054, 'TIME_S': 20.750195264816284, 'TIME_S_1KI': 2.560172148650991, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 630.7465198040009, 'W': 24.92526992228312}
[19.92, 20.0, 20.36, 20.36, 20.64, 20.92, 21.12, 21.0, 20.76, 20.36, 20.04, 20.08, 20.08, 20.4, 20.6, 20.76, 20.76, 20.76, 21.0, 20.84]
370.18
18.509
{'CPU': 'Altra', 'CORES': 80, 'ITERATIONS': 8105, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_120', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 106510, 'MATRIX_DENSITY': 0.0001081713341839054, 'TIME_S': 20.750195264816284, 'TIME_S_1KI': 2.560172148650991, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 630.7465198040009, 'W': 24.92526992228312, 'J_1KI': 77.8219025051204, 'W_1KI': 3.075295487018275, 'W_D': 6.416269922283121, 'J_D': 162.3669447200298, 'W_D_1KI': 0.7916434203927355, 'J_D_1KI': 0.09767346334271874}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_005.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 276410, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 25.263006687164307, "TIME_S_1KI": 0.09139686222337942, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2936.388348619938, "W": 107.27, "J_1KI": 10.623307219781983, "W_1KI": 0.388082920299555, "W_D": 70.633, "J_D": 1933.494157062292, "W_D_1KI": 0.2555370645056257, "J_D_1KI": 0.0009244855993112612}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_005.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_005.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 0.09953641891479492}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.9728, 0.9533, 0.8589, ..., 0.8433, 0.2350, 0.3886])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 0.09953641891479492 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '210978', '-m', 'matrices/as-caida_pruned/as-caida_G_005.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 16.028836250305176}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.0730, 0.6524, 0.0309, ..., 0.2548, 0.9366, 0.1429])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 16.028836250305176 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '276410', '-m', 'matrices/as-caida_pruned/as-caida_G_005.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 25.263006687164307}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.8011, 0.5871, 0.2879, ..., 0.3075, 0.3462, 0.5441])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 25.263006687164307 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.8011, 0.5871, 0.2879, ..., 0.3075, 0.3462, 0.5441])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 25.263006687164307 seconds
[40.1, 39.29, 39.65, 39.22, 39.19, 39.86, 39.23, 39.23, 39.15, 39.14]
[107.27]
27.373807668685913
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 276410, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_005', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 70026, 'MATRIX_DENSITY': 7.111825976492498e-05, 'TIME_S': 25.263006687164307, 'TIME_S_1KI': 0.09139686222337942, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2936.388348619938, 'W': 107.27}
[40.1, 39.29, 39.65, 39.22, 39.19, 39.86, 39.23, 39.23, 39.15, 39.14, 39.83, 39.44, 41.31, 44.88, 39.26, 39.36, 55.72, 39.24, 39.62, 39.11]
732.74
36.637
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 276410, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_005', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 70026, 'MATRIX_DENSITY': 7.111825976492498e-05, 'TIME_S': 25.263006687164307, 'TIME_S_1KI': 0.09139686222337942, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2936.388348619938, 'W': 107.27, 'J_1KI': 10.623307219781983, 'W_1KI': 0.388082920299555, 'W_D': 70.633, 'J_D': 1933.494157062292, 'W_D_1KI': 0.2555370645056257, 'J_D_1KI': 0.0009244855993112612}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_010.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 203597, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 21.505177974700928, "TIME_S_1KI": 0.10562620261939482, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2116.265174906254, "W": 105.79, "J_1KI": 10.394382898108782, "W_1KI": 0.5196049057697314, "W_D": 69.99100000000001, "J_D": 1400.1277611954215, "W_D_1KI": 0.3437722559762669, "J_D_1KI": 0.0016884937203213552}

65
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_010.output Normal file
View File

@ -0,0 +1,65 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_010.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 0.10314488410949707}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.5444, 0.3450, 0.0743, ..., 0.3559, 0.8180, 0.5488])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 0.10314488410949707 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '203597', '-m', 'matrices/as-caida_pruned/as-caida_G_010.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 21.505177974700928}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.1211, 0.6988, 0.4577, ..., 0.8664, 0.9808, 0.0173])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 21.505177974700928 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.1211, 0.6988, 0.4577, ..., 0.8664, 0.9808, 0.0173])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 21.505177974700928 seconds
[40.77, 40.16, 39.31, 39.46, 39.4, 39.45, 39.01, 40.43, 39.43, 39.2]
[105.79]
20.00439715385437
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 203597, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_010', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 74994, 'MATRIX_DENSITY': 7.616375021864427e-05, 'TIME_S': 21.505177974700928, 'TIME_S_1KI': 0.10562620261939482, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2116.265174906254, 'W': 105.79}
[40.77, 40.16, 39.31, 39.46, 39.4, 39.45, 39.01, 40.43, 39.43, 39.2, 39.65, 39.75, 38.91, 45.15, 39.07, 39.19, 39.47, 39.23, 39.35, 38.8]
715.98
35.799
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 203597, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_010', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 74994, 'MATRIX_DENSITY': 7.616375021864427e-05, 'TIME_S': 21.505177974700928, 'TIME_S_1KI': 0.10562620261939482, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2116.265174906254, 'W': 105.79, 'J_1KI': 10.394382898108782, 'W_1KI': 0.5196049057697314, 'W_D': 69.99100000000001, 'J_D': 1400.1277611954215, 'W_D_1KI': 0.3437722559762669, 'J_D_1KI': 0.0016884937203213552}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_015.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 238527, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_015", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 77124, "MATRIX_DENSITY": 7.832697378273889e-05, "TIME_S": 21.601000785827637, "TIME_S_1KI": 0.0905599818294266, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2453.5633573436735, "W": 107.46999999999998, "J_1KI": 10.286312901028703, "W_1KI": 0.4505569600087201, "W_D": 71.32874999999999, "J_D": 1628.4507985961434, "W_D_1KI": 0.2990384736319158, "J_D_1KI": 0.0012536881511607315}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_015.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_015.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_015", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 77124, "MATRIX_DENSITY": 7.832697378273889e-05, "TIME_S": 0.11880779266357422}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 4, 4, ..., 77124, 77124, 77124]),
col_indices=tensor([1040, 2054, 4842, ..., 160, 160, 8230]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=77124, layout=torch.sparse_csr)
tensor([0.1155, 0.9087, 0.1286, ..., 0.0633, 0.1640, 0.2660])
Matrix Type: SuiteSparse
Matrix: as-caida_G_015
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 77124
Density: 7.832697378273889e-05
Time: 0.11880779266357422 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '176756', '-m', 'matrices/as-caida_pruned/as-caida_G_015.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_015", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 77124, "MATRIX_DENSITY": 7.832697378273889e-05, "TIME_S": 15.56164002418518}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 4, 4, ..., 77124, 77124, 77124]),
col_indices=tensor([1040, 2054, 4842, ..., 160, 160, 8230]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=77124, layout=torch.sparse_csr)
tensor([0.1012, 0.9992, 0.7546, ..., 0.1559, 0.7607, 0.7528])
Matrix Type: SuiteSparse
Matrix: as-caida_G_015
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 77124
Density: 7.832697378273889e-05
Time: 15.56164002418518 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '238527', '-m', 'matrices/as-caida_pruned/as-caida_G_015.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_015", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 77124, "MATRIX_DENSITY": 7.832697378273889e-05, "TIME_S": 21.601000785827637}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 4, 4, ..., 77124, 77124, 77124]),
col_indices=tensor([1040, 2054, 4842, ..., 160, 160, 8230]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=77124, layout=torch.sparse_csr)
tensor([0.3942, 0.0726, 0.4805, ..., 0.0241, 0.4217, 0.4391])
Matrix Type: SuiteSparse
Matrix: as-caida_G_015
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 77124
Density: 7.832697378273889e-05
Time: 21.601000785827637 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 4, 4, ..., 77124, 77124, 77124]),
col_indices=tensor([1040, 2054, 4842, ..., 160, 160, 8230]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=77124, layout=torch.sparse_csr)
tensor([0.3942, 0.0726, 0.4805, ..., 0.0241, 0.4217, 0.4391])
Matrix Type: SuiteSparse
Matrix: as-caida_G_015
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 77124
Density: 7.832697378273889e-05
Time: 21.601000785827637 seconds
[40.23, 39.56, 39.22, 39.64, 39.02, 39.24, 39.06, 39.38, 39.16, 39.4]
[107.47]
22.83021640777588
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 238527, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_015', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 77124, 'MATRIX_DENSITY': 7.832697378273889e-05, 'TIME_S': 21.601000785827637, 'TIME_S_1KI': 0.0905599818294266, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2453.5633573436735, 'W': 107.46999999999998}
[40.23, 39.56, 39.22, 39.64, 39.02, 39.24, 39.06, 39.38, 39.16, 39.4, 41.7, 38.99, 39.59, 53.73, 39.02, 39.01, 39.05, 38.86, 39.89, 39.48]
722.825
36.14125
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 238527, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_015', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 77124, 'MATRIX_DENSITY': 7.832697378273889e-05, 'TIME_S': 21.601000785827637, 'TIME_S_1KI': 0.0905599818294266, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2453.5633573436735, 'W': 107.46999999999998, 'J_1KI': 10.286312901028703, 'W_1KI': 0.4505569600087201, 'W_D': 71.32874999999999, 'J_D': 1628.4507985961434, 'W_D_1KI': 0.2990384736319158, 'J_D_1KI': 0.0012536881511607315}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_020.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 252661, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_020", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 80948, "MATRIX_DENSITY": 8.221062021893506e-05, "TIME_S": 22.093090534210205, "TIME_S_1KI": 0.08744163339102673, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2827.9676496505735, "W": 106.8, "J_1KI": 11.19273512592198, "W_1KI": 0.42270077297248093, "W_D": 71.72475, "J_D": 1899.2066730269194, "W_D_1KI": 0.28387740886009316, "J_D_1KI": 0.0011235505632451908}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_020.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_020.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_020", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 80948, "MATRIX_DENSITY": 8.221062021893506e-05, "TIME_S": 0.10638713836669922}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 80944, 80946, 80948]),
col_indices=tensor([ 1040, 5699, 106, ..., 31378, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=80948, layout=torch.sparse_csr)
tensor([0.7616, 0.6085, 0.4474, ..., 0.0727, 0.0513, 0.0850])
Matrix Type: SuiteSparse
Matrix: as-caida_G_020
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 80948
Density: 8.221062021893506e-05
Time: 0.10638713836669922 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '197392', '-m', 'matrices/as-caida_pruned/as-caida_G_020.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_020", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 80948, "MATRIX_DENSITY": 8.221062021893506e-05, "TIME_S": 16.40624761581421}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 80944, 80946, 80948]),
col_indices=tensor([ 1040, 5699, 106, ..., 31378, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=80948, layout=torch.sparse_csr)
tensor([0.3330, 0.2122, 0.6427, ..., 0.2808, 0.4380, 0.9268])
Matrix Type: SuiteSparse
Matrix: as-caida_G_020
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 80948
Density: 8.221062021893506e-05
Time: 16.40624761581421 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '252661', '-m', 'matrices/as-caida_pruned/as-caida_G_020.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_020", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 80948, "MATRIX_DENSITY": 8.221062021893506e-05, "TIME_S": 22.093090534210205}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 80944, 80946, 80948]),
col_indices=tensor([ 1040, 5699, 106, ..., 31378, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=80948, layout=torch.sparse_csr)
tensor([0.8862, 0.2712, 0.1599, ..., 0.5989, 0.7417, 0.3691])
Matrix Type: SuiteSparse
Matrix: as-caida_G_020
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 80948
Density: 8.221062021893506e-05
Time: 22.093090534210205 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 80944, 80946, 80948]),
col_indices=tensor([ 1040, 5699, 106, ..., 31378, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=80948, layout=torch.sparse_csr)
tensor([0.8862, 0.2712, 0.1599, ..., 0.5989, 0.7417, 0.3691])
Matrix Type: SuiteSparse
Matrix: as-caida_G_020
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 80948
Density: 8.221062021893506e-05
Time: 22.093090534210205 seconds
[39.71, 38.73, 38.66, 38.74, 38.71, 38.81, 39.09, 38.65, 39.04, 38.91]
[106.8]
26.479097843170166
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 252661, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_020', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 80948, 'MATRIX_DENSITY': 8.221062021893506e-05, 'TIME_S': 22.093090534210205, 'TIME_S_1KI': 0.08744163339102673, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2827.9676496505735, 'W': 106.8}
[39.71, 38.73, 38.66, 38.74, 38.71, 38.81, 39.09, 38.65, 39.04, 38.91, 39.78, 38.72, 39.56, 38.58, 39.39, 39.1, 38.77, 39.29, 39.08, 38.77]
701.505
35.07525
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 252661, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_020', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 80948, 'MATRIX_DENSITY': 8.221062021893506e-05, 'TIME_S': 22.093090534210205, 'TIME_S_1KI': 0.08744163339102673, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2827.9676496505735, 'W': 106.8, 'J_1KI': 11.19273512592198, 'W_1KI': 0.42270077297248093, 'W_D': 71.72475, 'J_D': 1899.2066730269194, 'W_D_1KI': 0.28387740886009316, 'J_D_1KI': 0.0011235505632451908}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_025.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 240875, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_025", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 85850, "MATRIX_DENSITY": 8.718908121010495e-05, "TIME_S": 48.700708627700806, "TIME_S_1KI": 0.20218249560021093, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2510.559543380737, "W": 108.38, "J_1KI": 10.422665462919511, "W_1KI": 0.4499429164504411, "W_D": 72.928, "J_D": 1689.3346224365234, "W_D_1KI": 0.3027628437986507, "J_D_1KI": 0.0012569292944417257}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_025.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_025.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_025", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 85850, "MATRIX_DENSITY": 8.718908121010495e-05, "TIME_S": 2.1525959968566895}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 3, 3, ..., 85845, 85847, 85850]),
col_indices=tensor([ 346, 13811, 21783, ..., 15310, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=85850, layout=torch.sparse_csr)
tensor([0.1548, 0.6807, 0.1439, ..., 0.0147, 0.6028, 0.1765])
Matrix Type: SuiteSparse
Matrix: as-caida_G_025
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 85850
Density: 8.718908121010495e-05
Time: 2.1525959968566895 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '9755', '-m', 'matrices/as-caida_pruned/as-caida_G_025.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_025", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 85850, "MATRIX_DENSITY": 8.718908121010495e-05, "TIME_S": 0.850459098815918}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 3, 3, ..., 85845, 85847, 85850]),
col_indices=tensor([ 346, 13811, 21783, ..., 15310, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=85850, layout=torch.sparse_csr)
tensor([0.3242, 0.6133, 0.1194, ..., 0.5981, 0.5897, 0.0971])
Matrix Type: SuiteSparse
Matrix: as-caida_G_025
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 85850
Density: 8.718908121010495e-05
Time: 0.850459098815918 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '240875', '-m', 'matrices/as-caida_pruned/as-caida_G_025.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_025", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 85850, "MATRIX_DENSITY": 8.718908121010495e-05, "TIME_S": 48.700708627700806}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 3, 3, ..., 85845, 85847, 85850]),
col_indices=tensor([ 346, 13811, 21783, ..., 15310, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=85850, layout=torch.sparse_csr)
tensor([0.6774, 0.9329, 0.5853, ..., 0.8274, 0.5489, 0.8291])
Matrix Type: SuiteSparse
Matrix: as-caida_G_025
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 85850
Density: 8.718908121010495e-05
Time: 48.700708627700806 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 3, 3, ..., 85845, 85847, 85850]),
col_indices=tensor([ 346, 13811, 21783, ..., 15310, 17998, 31377]),
values=tensor([1., 1., 1., ..., 1., 1., 3.]), size=(31379, 31379),
nnz=85850, layout=torch.sparse_csr)
tensor([0.6774, 0.9329, 0.5853, ..., 0.8274, 0.5489, 0.8291])
Matrix Type: SuiteSparse
Matrix: as-caida_G_025
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 85850
Density: 8.718908121010495e-05
Time: 48.700708627700806 seconds
[40.43, 39.93, 39.14, 39.8, 38.99, 39.48, 39.06, 38.95, 39.29, 38.89]
[108.38]
23.164417266845703
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 240875, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_025', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 85850, 'MATRIX_DENSITY': 8.718908121010495e-05, 'TIME_S': 48.700708627700806, 'TIME_S_1KI': 0.20218249560021093, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2510.559543380737, 'W': 108.38}
[40.43, 39.93, 39.14, 39.8, 38.99, 39.48, 39.06, 38.95, 39.29, 38.89, 40.82, 39.02, 39.15, 39.35, 39.01, 39.32, 39.03, 39.3, 39.12, 42.06]
709.04
35.452
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 240875, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_025', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 85850, 'MATRIX_DENSITY': 8.718908121010495e-05, 'TIME_S': 48.700708627700806, 'TIME_S_1KI': 0.20218249560021093, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2510.559543380737, 'W': 108.38, 'J_1KI': 10.422665462919511, 'W_1KI': 0.4499429164504411, 'W_D': 72.928, 'J_D': 1689.3346224365234, 'W_D_1KI': 0.3027628437986507, 'J_D_1KI': 0.0012569292944417257}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_030.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 245946, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_030", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 86850, "MATRIX_DENSITY": 8.820467912752026e-05, "TIME_S": 21.173834562301636, "TIME_S_1KI": 0.08609139633212833, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2654.1946601867676, "W": 108.5, "J_1KI": 10.791778114654305, "W_1KI": 0.4411537491969782, "W_D": 73.0335, "J_D": 1786.5910204124452, "W_D_1KI": 0.29694933034080656, "J_D_1KI": 0.001207376132731602}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_030.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_030.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_030", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 86850, "MATRIX_DENSITY": 8.820467912752026e-05, "TIME_S": 0.11014795303344727}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 86850, 86850, 86850]),
col_indices=tensor([ 1809, 21783, 106, ..., 7018, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=86850, layout=torch.sparse_csr)
tensor([0.7332, 0.7503, 0.1338, ..., 0.4921, 0.4099, 0.1438])
Matrix Type: SuiteSparse
Matrix: as-caida_G_030
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 86850
Density: 8.820467912752026e-05
Time: 0.11014795303344727 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '190652', '-m', 'matrices/as-caida_pruned/as-caida_G_030.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_030", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 86850, "MATRIX_DENSITY": 8.820467912752026e-05, "TIME_S": 16.278696298599243}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 86850, 86850, 86850]),
col_indices=tensor([ 1809, 21783, 106, ..., 7018, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=86850, layout=torch.sparse_csr)
tensor([0.4532, 0.7021, 0.2379, ..., 0.4729, 0.4308, 0.9482])
Matrix Type: SuiteSparse
Matrix: as-caida_G_030
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 86850
Density: 8.820467912752026e-05
Time: 16.278696298599243 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '245946', '-m', 'matrices/as-caida_pruned/as-caida_G_030.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_030", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 86850, "MATRIX_DENSITY": 8.820467912752026e-05, "TIME_S": 21.173834562301636}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 86850, 86850, 86850]),
col_indices=tensor([ 1809, 21783, 106, ..., 7018, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=86850, layout=torch.sparse_csr)
tensor([0.3998, 0.0719, 0.5574, ..., 0.0984, 0.7335, 0.9981])
Matrix Type: SuiteSparse
Matrix: as-caida_G_030
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 86850
Density: 8.820467912752026e-05
Time: 21.173834562301636 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 86850, 86850, 86850]),
col_indices=tensor([ 1809, 21783, 106, ..., 7018, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=86850, layout=torch.sparse_csr)
tensor([0.3998, 0.0719, 0.5574, ..., 0.0984, 0.7335, 0.9981])
Matrix Type: SuiteSparse
Matrix: as-caida_G_030
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 86850
Density: 8.820467912752026e-05
Time: 21.173834562301636 seconds
[40.2, 39.33, 39.22, 39.59, 39.19, 39.41, 39.13, 39.18, 39.7, 39.19]
[108.5]
24.462623596191406
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 245946, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_030', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 86850, 'MATRIX_DENSITY': 8.820467912752026e-05, 'TIME_S': 21.173834562301636, 'TIME_S_1KI': 0.08609139633212833, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2654.1946601867676, 'W': 108.5}
[40.2, 39.33, 39.22, 39.59, 39.19, 39.41, 39.13, 39.18, 39.7, 39.19, 39.75, 39.67, 39.11, 39.3, 39.17, 39.45, 39.49, 39.54, 39.42, 39.72]
709.3299999999999
35.466499999999996
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 245946, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_030', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 86850, 'MATRIX_DENSITY': 8.820467912752026e-05, 'TIME_S': 21.173834562301636, 'TIME_S_1KI': 0.08609139633212833, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2654.1946601867676, 'W': 108.5, 'J_1KI': 10.791778114654305, 'W_1KI': 0.4411537491969782, 'W_D': 73.0335, 'J_D': 1786.5910204124452, 'W_D_1KI': 0.29694933034080656, 'J_D_1KI': 0.001207376132731602}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_035.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 235955, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_035", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 87560, "MATRIX_DENSITY": 8.892575364888514e-05, "TIME_S": 21.57011389732361, "TIME_S_1KI": 0.09141621875918547, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2532.983486351967, "W": 108.50999999999999, "J_1KI": 10.735027807641146, "W_1KI": 0.459875823779958, "W_D": 72.94724999999998, "J_D": 1702.8308877042527, "W_D_1KI": 0.3091574664660634, "J_D_1KI": 0.0013102390984131016}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_035.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_035.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_035", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 87560, "MATRIX_DENSITY": 8.892575364888514e-05, "TIME_S": 0.10548853874206543}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 87559, 87559, 87560]),
col_indices=tensor([ 1809, 21783, 106, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=87560, layout=torch.sparse_csr)
tensor([0.7539, 0.5110, 0.4867, ..., 0.4276, 0.9205, 0.2155])
Matrix Type: SuiteSparse
Matrix: as-caida_G_035
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 87560
Density: 8.892575364888514e-05
Time: 0.10548853874206543 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '199073', '-m', 'matrices/as-caida_pruned/as-caida_G_035.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_035", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 87560, "MATRIX_DENSITY": 8.892575364888514e-05, "TIME_S": 17.71744680404663}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 87559, 87559, 87560]),
col_indices=tensor([ 1809, 21783, 106, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=87560, layout=torch.sparse_csr)
tensor([0.3160, 0.4163, 0.7745, ..., 0.9698, 0.5064, 0.9398])
Matrix Type: SuiteSparse
Matrix: as-caida_G_035
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 87560
Density: 8.892575364888514e-05
Time: 17.71744680404663 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '235955', '-m', 'matrices/as-caida_pruned/as-caida_G_035.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_035", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 87560, "MATRIX_DENSITY": 8.892575364888514e-05, "TIME_S": 21.57011389732361}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 87559, 87559, 87560]),
col_indices=tensor([ 1809, 21783, 106, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=87560, layout=torch.sparse_csr)
tensor([0.4189, 0.3756, 0.7284, ..., 0.6039, 0.3176, 0.7868])
Matrix Type: SuiteSparse
Matrix: as-caida_G_035
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 87560
Density: 8.892575364888514e-05
Time: 21.57011389732361 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 87559, 87559, 87560]),
col_indices=tensor([ 1809, 21783, 106, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=87560, layout=torch.sparse_csr)
tensor([0.4189, 0.3756, 0.7284, ..., 0.6039, 0.3176, 0.7868])
Matrix Type: SuiteSparse
Matrix: as-caida_G_035
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 87560
Density: 8.892575364888514e-05
Time: 21.57011389732361 seconds
[40.52, 39.16, 39.66, 39.15, 39.15, 39.09, 39.3, 39.32, 39.15, 44.41]
[108.51]
23.343318462371826
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 235955, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_035', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 87560, 'MATRIX_DENSITY': 8.892575364888514e-05, 'TIME_S': 21.57011389732361, 'TIME_S_1KI': 0.09141621875918547, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2532.983486351967, 'W': 108.50999999999999}
[40.52, 39.16, 39.66, 39.15, 39.15, 39.09, 39.3, 39.32, 39.15, 44.41, 39.78, 39.5, 39.15, 38.99, 39.53, 38.97, 39.96, 38.99, 40.12, 39.42]
711.2550000000001
35.56275000000001
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 235955, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_035', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 87560, 'MATRIX_DENSITY': 8.892575364888514e-05, 'TIME_S': 21.57011389732361, 'TIME_S_1KI': 0.09141621875918547, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2532.983486351967, 'W': 108.50999999999999, 'J_1KI': 10.735027807641146, 'W_1KI': 0.459875823779958, 'W_D': 72.94724999999998, 'J_D': 1702.8308877042527, 'W_D_1KI': 0.3091574664660634, 'J_D_1KI': 0.0013102390984131016}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_040.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 246041, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_040", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89658, "MATRIX_DENSITY": 9.105647807962247e-05, "TIME_S": 21.844791889190674, "TIME_S_1KI": 0.08878516950098021, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2588.3357152462004, "W": 108.77999999999999, "J_1KI": 10.519936576612029, "W_1KI": 0.44212143504537854, "W_D": 73.46974999999999, "J_D": 1748.155707990527, "W_D_1KI": 0.29860775236647547, "J_D_1KI": 0.0012136503768334361}

105
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_040.output Normal file
View File

@ -0,0 +1,105 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_040.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_040", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89658, "MATRIX_DENSITY": 9.105647807962247e-05, "TIME_S": 0.10488057136535645}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89657, 89657, 89658]),
col_indices=tensor([ 106, 329, 1040, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89658, layout=torch.sparse_csr)
tensor([0.6314, 0.4358, 0.9289, ..., 0.8653, 0.8137, 0.1981])
Matrix Type: SuiteSparse
Matrix: as-caida_G_040
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89658
Density: 9.105647807962247e-05
Time: 0.10488057136535645 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '200227', '-m', 'matrices/as-caida_pruned/as-caida_G_040.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_040", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89658, "MATRIX_DENSITY": 9.105647807962247e-05, "TIME_S": 18.238918781280518}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89657, 89657, 89658]),
col_indices=tensor([ 106, 329, 1040, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89658, layout=torch.sparse_csr)
tensor([0.2930, 0.0363, 0.8927, ..., 0.0958, 0.8570, 0.3046])
Matrix Type: SuiteSparse
Matrix: as-caida_G_040
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89658
Density: 9.105647807962247e-05
Time: 18.238918781280518 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '230538', '-m', 'matrices/as-caida_pruned/as-caida_G_040.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_040", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89658, "MATRIX_DENSITY": 9.105647807962247e-05, "TIME_S": 19.67674446105957}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89657, 89657, 89658]),
col_indices=tensor([ 106, 329, 1040, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89658, layout=torch.sparse_csr)
tensor([0.6913, 0.2101, 0.6607, ..., 0.9927, 0.2277, 0.7104])
Matrix Type: SuiteSparse
Matrix: as-caida_G_040
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89658
Density: 9.105647807962247e-05
Time: 19.67674446105957 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '246041', '-m', 'matrices/as-caida_pruned/as-caida_G_040.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_040", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89658, "MATRIX_DENSITY": 9.105647807962247e-05, "TIME_S": 21.844791889190674}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89657, 89657, 89658]),
col_indices=tensor([ 106, 329, 1040, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89658, layout=torch.sparse_csr)
tensor([0.8541, 0.2196, 0.7858, ..., 0.4806, 0.8499, 0.4712])
Matrix Type: SuiteSparse
Matrix: as-caida_G_040
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89658
Density: 9.105647807962247e-05
Time: 21.844791889190674 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89657, 89657, 89658]),
col_indices=tensor([ 106, 329, 1040, ..., 10144, 882, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89658, layout=torch.sparse_csr)
tensor([0.8541, 0.2196, 0.7858, ..., 0.4806, 0.8499, 0.4712])
Matrix Type: SuiteSparse
Matrix: as-caida_G_040
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89658
Density: 9.105647807962247e-05
Time: 21.844791889190674 seconds
[40.03, 38.91, 40.66, 39.33, 39.14, 38.85, 39.49, 39.08, 39.33, 38.84]
[108.78]
23.79422426223755
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 246041, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_040', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 89658, 'MATRIX_DENSITY': 9.105647807962247e-05, 'TIME_S': 21.844791889190674, 'TIME_S_1KI': 0.08878516950098021, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2588.3357152462004, 'W': 108.77999999999999}
[40.03, 38.91, 40.66, 39.33, 39.14, 38.85, 39.49, 39.08, 39.33, 38.84, 39.77, 38.86, 39.28, 38.9, 38.94, 38.78, 38.83, 39.29, 39.53, 39.37]
706.2049999999999
35.310249999999996
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 246041, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_040', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 89658, 'MATRIX_DENSITY': 9.105647807962247e-05, 'TIME_S': 21.844791889190674, 'TIME_S_1KI': 0.08878516950098021, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2588.3357152462004, 'W': 108.77999999999999, 'J_1KI': 10.519936576612029, 'W_1KI': 0.44212143504537854, 'W_D': 73.46974999999999, 'J_D': 1748.155707990527, 'W_D_1KI': 0.29860775236647547, 'J_D_1KI': 0.0012136503768334361}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_045.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 243173, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_045", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89152, "MATRIX_DENSITY": 9.054258553341032e-05, "TIME_S": 21.187162399291992, "TIME_S_1KI": 0.08712793936535713, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2563.625514292717, "W": 108.94, "J_1KI": 10.542393745575032, "W_1KI": 0.447993815102828, "W_D": 73.28999999999999, "J_D": 1724.693537199497, "W_D_1KI": 0.30139036817409826, "J_D_1KI": 0.0012394072046407218}

105
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_045.output Normal file
View File

@ -0,0 +1,105 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_045.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_045", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89152, "MATRIX_DENSITY": 9.054258553341032e-05, "TIME_S": 0.10718226432800293}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89150, 89150, 89152]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89152, layout=torch.sparse_csr)
tensor([0.0096, 0.2145, 0.6490, ..., 0.2282, 0.3485, 0.9521])
Matrix Type: SuiteSparse
Matrix: as-caida_G_045
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89152
Density: 9.054258553341032e-05
Time: 0.10718226432800293 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '195927', '-m', 'matrices/as-caida_pruned/as-caida_G_045.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_045", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89152, "MATRIX_DENSITY": 9.054258553341032e-05, "TIME_S": 18.098947763442993}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89150, 89150, 89152]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89152, layout=torch.sparse_csr)
tensor([0.5756, 0.6430, 0.3728, ..., 0.7554, 0.9060, 0.8860])
Matrix Type: SuiteSparse
Matrix: as-caida_G_045
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89152
Density: 9.054258553341032e-05
Time: 18.098947763442993 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '227331', '-m', 'matrices/as-caida_pruned/as-caida_G_045.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_045", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89152, "MATRIX_DENSITY": 9.054258553341032e-05, "TIME_S": 19.63188910484314}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89150, 89150, 89152]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89152, layout=torch.sparse_csr)
tensor([0.9737, 0.4514, 0.8059, ..., 0.3086, 0.9830, 0.5037])
Matrix Type: SuiteSparse
Matrix: as-caida_G_045
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89152
Density: 9.054258553341032e-05
Time: 19.63188910484314 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '243173', '-m', 'matrices/as-caida_pruned/as-caida_G_045.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_045", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 89152, "MATRIX_DENSITY": 9.054258553341032e-05, "TIME_S": 21.187162399291992}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89150, 89150, 89152]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89152, layout=torch.sparse_csr)
tensor([0.5566, 0.0730, 0.0280, ..., 0.4513, 0.9009, 0.5894])
Matrix Type: SuiteSparse
Matrix: as-caida_G_045
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89152
Density: 9.054258553341032e-05
Time: 21.187162399291992 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 89150, 89150, 89152]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=89152, layout=torch.sparse_csr)
tensor([0.5566, 0.0730, 0.0280, ..., 0.4513, 0.9009, 0.5894])
Matrix Type: SuiteSparse
Matrix: as-caida_G_045
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 89152
Density: 9.054258553341032e-05
Time: 21.187162399291992 seconds
[39.8, 39.71, 39.11, 39.44, 39.1, 39.43, 39.16, 39.54, 39.55, 39.14]
[108.94]
23.532453775405884
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 243173, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_045', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 89152, 'MATRIX_DENSITY': 9.054258553341032e-05, 'TIME_S': 21.187162399291992, 'TIME_S_1KI': 0.08712793936535713, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2563.625514292717, 'W': 108.94}
[39.8, 39.71, 39.11, 39.44, 39.1, 39.43, 39.16, 39.54, 39.55, 39.14, 39.82, 39.18, 39.5, 39.07, 38.94, 39.22, 39.0, 44.95, 38.94, 39.56]
713.0
35.65
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 243173, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_045', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 89152, 'MATRIX_DENSITY': 9.054258553341032e-05, 'TIME_S': 21.187162399291992, 'TIME_S_1KI': 0.08712793936535713, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2563.625514292717, 'W': 108.94, 'J_1KI': 10.542393745575032, 'W_1KI': 0.447993815102828, 'W_D': 73.28999999999999, 'J_D': 1724.693537199497, 'W_D_1KI': 0.30139036817409826, 'J_D_1KI': 0.0012394072046407218}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_050.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 240158, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_050", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 90392, "MATRIX_DENSITY": 9.180192695100532e-05, "TIME_S": 22.33744502067566, "TIME_S_1KI": 0.09301145504491068, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2562.788330435753, "W": 107.15, "J_1KI": 10.67125946433495, "W_1KI": 0.44616460829953614, "W_D": 72.1175, "J_D": 1724.889289969206, "W_D_1KI": 0.3002918911716454, "J_D_1KI": 0.0012503930377986384}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_050.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_050.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_050", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 90392, "MATRIX_DENSITY": 9.180192695100532e-05, "TIME_S": 0.13907456398010254}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 90390, 90390, 90392]),
col_indices=tensor([ 5326, 106, 329, ..., 882, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=90392, layout=torch.sparse_csr)
tensor([0.7147, 0.9892, 0.5644, ..., 0.7484, 0.7920, 0.9091])
Matrix Type: SuiteSparse
Matrix: as-caida_G_050
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 90392
Density: 9.180192695100532e-05
Time: 0.13907456398010254 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '150998', '-m', 'matrices/as-caida_pruned/as-caida_G_050.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_050", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 90392, "MATRIX_DENSITY": 9.180192695100532e-05, "TIME_S": 13.203584909439087}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 90390, 90390, 90392]),
col_indices=tensor([ 5326, 106, 329, ..., 882, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=90392, layout=torch.sparse_csr)
tensor([0.4141, 0.2334, 0.4747, ..., 0.7663, 0.6533, 0.8550])
Matrix Type: SuiteSparse
Matrix: as-caida_G_050
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 90392
Density: 9.180192695100532e-05
Time: 13.203584909439087 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '240158', '-m', 'matrices/as-caida_pruned/as-caida_G_050.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_050", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 90392, "MATRIX_DENSITY": 9.180192695100532e-05, "TIME_S": 22.33744502067566}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 90390, 90390, 90392]),
col_indices=tensor([ 5326, 106, 329, ..., 882, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=90392, layout=torch.sparse_csr)
tensor([0.5149, 0.8136, 0.7771, ..., 0.9064, 0.0587, 0.9445])
Matrix Type: SuiteSparse
Matrix: as-caida_G_050
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 90392
Density: 9.180192695100532e-05
Time: 22.33744502067566 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 90390, 90390, 90392]),
col_indices=tensor([ 5326, 106, 329, ..., 882, 2232, 16085]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=90392, layout=torch.sparse_csr)
tensor([0.5149, 0.8136, 0.7771, ..., 0.9064, 0.0587, 0.9445])
Matrix Type: SuiteSparse
Matrix: as-caida_G_050
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 90392
Density: 9.180192695100532e-05
Time: 22.33744502067566 seconds
[44.15, 39.09, 38.82, 38.46, 38.4, 38.77, 38.41, 38.29, 38.84, 38.75]
[107.15]
23.917763233184814
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 240158, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_050', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 90392, 'MATRIX_DENSITY': 9.180192695100532e-05, 'TIME_S': 22.33744502067566, 'TIME_S_1KI': 0.09301145504491068, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2562.788330435753, 'W': 107.15}
[44.15, 39.09, 38.82, 38.46, 38.4, 38.77, 38.41, 38.29, 38.84, 38.75, 39.2, 39.86, 38.62, 38.82, 38.51, 38.9, 39.2, 38.9, 38.42, 38.58]
700.6500000000001
35.032500000000006
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 240158, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_050', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 90392, 'MATRIX_DENSITY': 9.180192695100532e-05, 'TIME_S': 22.33744502067566, 'TIME_S_1KI': 0.09301145504491068, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2562.788330435753, 'W': 107.15, 'J_1KI': 10.67125946433495, 'W_1KI': 0.44616460829953614, 'W_D': 72.1175, 'J_D': 1724.889289969206, 'W_D_1KI': 0.3002918911716454, 'J_D_1KI': 0.0012503930377986384}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_055.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 243056, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_055", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 91476, "MATRIX_DENSITY": 9.290283509348351e-05, "TIME_S": 21.788699865341187, "TIME_S_1KI": 0.0896447726669623, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2614.8328268265723, "W": 109.47, "J_1KI": 10.758149672612781, "W_1KI": 0.45039003357251, "W_D": 73.9525, "J_D": 1766.4513074439765, "W_D_1KI": 0.3042611579224541, "J_D_1KI": 0.0012518150464191549}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_055.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_055.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_055", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 91476, "MATRIX_DENSITY": 9.290283509348351e-05, "TIME_S": 0.12037158012390137}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 91475, 91475, 91476]),
col_indices=tensor([21783, 106, 329, ..., 160, 882, 17255]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=91476, layout=torch.sparse_csr)
tensor([0.2129, 0.8233, 0.4187, ..., 0.5468, 0.2620, 0.5585])
Matrix Type: SuiteSparse
Matrix: as-caida_G_055
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 91476
Density: 9.290283509348351e-05
Time: 0.12037158012390137 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '174459', '-m', 'matrices/as-caida_pruned/as-caida_G_055.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_055", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 91476, "MATRIX_DENSITY": 9.290283509348351e-05, "TIME_S": 15.073203563690186}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 91475, 91475, 91476]),
col_indices=tensor([21783, 106, 329, ..., 160, 882, 17255]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=91476, layout=torch.sparse_csr)
tensor([0.6340, 0.2808, 0.1161, ..., 0.6453, 0.0525, 0.0651])
Matrix Type: SuiteSparse
Matrix: as-caida_G_055
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 91476
Density: 9.290283509348351e-05
Time: 15.073203563690186 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '243056', '-m', 'matrices/as-caida_pruned/as-caida_G_055.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_055", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 91476, "MATRIX_DENSITY": 9.290283509348351e-05, "TIME_S": 21.788699865341187}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 91475, 91475, 91476]),
col_indices=tensor([21783, 106, 329, ..., 160, 882, 17255]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=91476, layout=torch.sparse_csr)
tensor([0.0218, 0.7016, 0.7341, ..., 0.8849, 0.1906, 0.8954])
Matrix Type: SuiteSparse
Matrix: as-caida_G_055
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 91476
Density: 9.290283509348351e-05
Time: 21.788699865341187 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 91475, 91475, 91476]),
col_indices=tensor([21783, 106, 329, ..., 160, 882, 17255]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=91476, layout=torch.sparse_csr)
tensor([0.0218, 0.7016, 0.7341, ..., 0.8849, 0.1906, 0.8954])
Matrix Type: SuiteSparse
Matrix: as-caida_G_055
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 91476
Density: 9.290283509348351e-05
Time: 21.788699865341187 seconds
[40.42, 39.66, 39.33, 39.53, 39.2, 39.11, 39.99, 39.13, 39.14, 39.51]
[109.47]
23.886296033859253
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 243056, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_055', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 91476, 'MATRIX_DENSITY': 9.290283509348351e-05, 'TIME_S': 21.788699865341187, 'TIME_S_1KI': 0.0896447726669623, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2614.8328268265723, 'W': 109.47}
[40.42, 39.66, 39.33, 39.53, 39.2, 39.11, 39.99, 39.13, 39.14, 39.51, 40.07, 39.67, 39.37, 39.27, 39.27, 39.47, 39.3, 39.93, 39.46, 39.04]
710.35
35.5175
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 243056, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_055', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 91476, 'MATRIX_DENSITY': 9.290283509348351e-05, 'TIME_S': 21.788699865341187, 'TIME_S_1KI': 0.0896447726669623, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2614.8328268265723, 'W': 109.47, 'J_1KI': 10.758149672612781, 'W_1KI': 0.45039003357251, 'W_D': 73.9525, 'J_D': 1766.4513074439765, 'W_D_1KI': 0.3042611579224541, 'J_D_1KI': 0.0012518150464191549}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_060.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 231053, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_060", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 94180, "MATRIX_DENSITY": 9.564901186217454e-05, "TIME_S": 20.448925495147705, "TIME_S_1KI": 0.08850318106732094, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2488.9935939955712, "W": 109.33, "J_1KI": 10.772392455391495, "W_1KI": 0.47318147784274606, "W_D": 73.588, "J_D": 1675.2955327444074, "W_D_1KI": 0.3184896971690478, "J_D_1KI": 0.001378427015312711}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_060.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_060.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_060", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 94180, "MATRIX_DENSITY": 9.564901186217454e-05, "TIME_S": 0.10647058486938477}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 94180, 94180, 94180]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=94180, layout=torch.sparse_csr)
tensor([0.3889, 0.3396, 0.4863, ..., 0.8323, 0.8906, 0.8333])
Matrix Type: SuiteSparse
Matrix: as-caida_G_060
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 94180
Density: 9.564901186217454e-05
Time: 0.10647058486938477 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '197237', '-m', 'matrices/as-caida_pruned/as-caida_G_060.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_060", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 94180, "MATRIX_DENSITY": 9.564901186217454e-05, "TIME_S": 17.926481008529663}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 94180, 94180, 94180]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=94180, layout=torch.sparse_csr)
tensor([0.0027, 0.6119, 0.0510, ..., 0.1297, 0.5480, 0.0399])
Matrix Type: SuiteSparse
Matrix: as-caida_G_060
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 94180
Density: 9.564901186217454e-05
Time: 17.926481008529663 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '231053', '-m', 'matrices/as-caida_pruned/as-caida_G_060.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_060", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 94180, "MATRIX_DENSITY": 9.564901186217454e-05, "TIME_S": 20.448925495147705}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 94180, 94180, 94180]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=94180, layout=torch.sparse_csr)
tensor([0.2183, 0.4339, 0.4856, ..., 0.4687, 0.5468, 0.7375])
Matrix Type: SuiteSparse
Matrix: as-caida_G_060
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 94180
Density: 9.564901186217454e-05
Time: 20.448925495147705 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 94180, 94180, 94180]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=94180, layout=torch.sparse_csr)
tensor([0.2183, 0.4339, 0.4856, ..., 0.4687, 0.5468, 0.7375])
Matrix Type: SuiteSparse
Matrix: as-caida_G_060
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 94180
Density: 9.564901186217454e-05
Time: 20.448925495147705 seconds
[41.56, 39.18, 39.21, 39.16, 39.44, 45.24, 39.35, 39.09, 39.61, 39.47]
[109.33]
22.7658793926239
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 231053, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_060', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 94180, 'MATRIX_DENSITY': 9.564901186217454e-05, 'TIME_S': 20.448925495147705, 'TIME_S_1KI': 0.08850318106732094, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2488.9935939955712, 'W': 109.33}
[41.56, 39.18, 39.21, 39.16, 39.44, 45.24, 39.35, 39.09, 39.61, 39.47, 40.17, 39.63, 39.12, 39.22, 39.1, 39.33, 39.13, 39.09, 39.61, 39.46]
714.8400000000001
35.742000000000004
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 231053, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_060', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 94180, 'MATRIX_DENSITY': 9.564901186217454e-05, 'TIME_S': 20.448925495147705, 'TIME_S_1KI': 0.08850318106732094, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2488.9935939955712, 'W': 109.33, 'J_1KI': 10.772392455391495, 'W_1KI': 0.47318147784274606, 'W_D': 73.588, 'J_D': 1675.2955327444074, 'W_D_1KI': 0.3184896971690478, 'J_D_1KI': 0.001378427015312711}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_065.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 251814, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_065", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 95068, "MATRIX_DENSITY": 9.655086281283934e-05, "TIME_S": 21.933937549591064, "TIME_S_1KI": 0.08710372556565983, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2850.223569984436, "W": 109.16, "J_1KI": 11.318765318784642, "W_1KI": 0.43349456344762405, "W_D": 73.41199999999999, "J_D": 1916.8249607887267, "W_D_1KI": 0.29153263917018113, "J_D_1KI": 0.0011577300673123064}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_065.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_065.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_065", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 95068, "MATRIX_DENSITY": 9.655086281283934e-05, "TIME_S": 0.1031339168548584}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 95068, 95068, 95068]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=95068, layout=torch.sparse_csr)
tensor([0.3467, 0.1204, 0.8088, ..., 0.7162, 0.4097, 0.3295])
Matrix Type: SuiteSparse
Matrix: as-caida_G_065
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 95068
Density: 9.655086281283934e-05
Time: 0.1031339168548584 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '203618', '-m', 'matrices/as-caida_pruned/as-caida_G_065.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_065", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 95068, "MATRIX_DENSITY": 9.655086281283934e-05, "TIME_S": 16.980651140213013}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 95068, 95068, 95068]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=95068, layout=torch.sparse_csr)
tensor([0.1656, 0.7627, 0.7205, ..., 0.6513, 0.9973, 0.1356])
Matrix Type: SuiteSparse
Matrix: as-caida_G_065
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 95068
Density: 9.655086281283934e-05
Time: 16.980651140213013 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '251814', '-m', 'matrices/as-caida_pruned/as-caida_G_065.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_065", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 95068, "MATRIX_DENSITY": 9.655086281283934e-05, "TIME_S": 21.933937549591064}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 95068, 95068, 95068]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=95068, layout=torch.sparse_csr)
tensor([0.6795, 0.4572, 0.8752, ..., 0.2627, 0.4989, 0.4040])
Matrix Type: SuiteSparse
Matrix: as-caida_G_065
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 95068
Density: 9.655086281283934e-05
Time: 21.933937549591064 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 95068, 95068, 95068]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 882]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=95068, layout=torch.sparse_csr)
tensor([0.6795, 0.4572, 0.8752, ..., 0.2627, 0.4989, 0.4040])
Matrix Type: SuiteSparse
Matrix: as-caida_G_065
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 95068
Density: 9.655086281283934e-05
Time: 21.933937549591064 seconds
[40.24, 39.25, 39.88, 39.1, 44.85, 39.19, 39.6, 39.39, 39.16, 39.14]
[109.16]
26.110512733459473
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 251814, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_065', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 95068, 'MATRIX_DENSITY': 9.655086281283934e-05, 'TIME_S': 21.933937549591064, 'TIME_S_1KI': 0.08710372556565983, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2850.223569984436, 'W': 109.16}
[40.24, 39.25, 39.88, 39.1, 44.85, 39.19, 39.6, 39.39, 39.16, 39.14, 39.92, 39.63, 39.78, 40.29, 39.17, 39.06, 39.09, 39.17, 39.2, 39.0]
714.96
35.748000000000005
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 251814, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_065', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 95068, 'MATRIX_DENSITY': 9.655086281283934e-05, 'TIME_S': 21.933937549591064, 'TIME_S_1KI': 0.08710372556565983, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2850.223569984436, 'W': 109.16, 'J_1KI': 11.318765318784642, 'W_1KI': 0.43349456344762405, 'W_D': 73.41199999999999, 'J_D': 1916.8249607887267, 'W_D_1KI': 0.29153263917018113, 'J_D_1KI': 0.0011577300673123064}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_070.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 246006, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_070", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 78684, "MATRIX_DENSITY": 7.991130653390679e-05, "TIME_S": 22.265681505203247, "TIME_S_1KI": 0.0905086928985604, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2452.5237454128264, "W": 108.92, "J_1KI": 9.969365565932645, "W_1KI": 0.4427534287781599, "W_D": 73.40100000000001, "J_D": 1652.751518885851, "W_D_1KI": 0.29837077144459895, "J_D_1KI": 0.0012128597328707386}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_070.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_070.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_070", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 78684, "MATRIX_DENSITY": 7.991130653390679e-05, "TIME_S": 0.10181021690368652}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 78684, 78684, 78684]),
col_indices=tensor([ 106, 329, 1040, ..., 16263, 2242, 2242]),
values=tensor([1., 1., 1., ..., 3., 1., 1.]), size=(31379, 31379),
nnz=78684, layout=torch.sparse_csr)
tensor([0.7820, 0.8357, 0.3400, ..., 0.6426, 0.8336, 0.2071])
Matrix Type: SuiteSparse
Matrix: as-caida_G_070
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 78684
Density: 7.991130653390679e-05
Time: 0.10181021690368652 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '206266', '-m', 'matrices/as-caida_pruned/as-caida_G_070.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_070", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 78684, "MATRIX_DENSITY": 7.991130653390679e-05, "TIME_S": 17.607593774795532}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 78684, 78684, 78684]),
col_indices=tensor([ 106, 329, 1040, ..., 16263, 2242, 2242]),
values=tensor([1., 1., 1., ..., 3., 1., 1.]), size=(31379, 31379),
nnz=78684, layout=torch.sparse_csr)
tensor([0.9459, 0.2541, 0.2751, ..., 0.1790, 0.8403, 0.0106])
Matrix Type: SuiteSparse
Matrix: as-caida_G_070
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 78684
Density: 7.991130653390679e-05
Time: 17.607593774795532 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '246006', '-m', 'matrices/as-caida_pruned/as-caida_G_070.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_070", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 78684, "MATRIX_DENSITY": 7.991130653390679e-05, "TIME_S": 22.265681505203247}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 78684, 78684, 78684]),
col_indices=tensor([ 106, 329, 1040, ..., 16263, 2242, 2242]),
values=tensor([1., 1., 1., ..., 3., 1., 1.]), size=(31379, 31379),
nnz=78684, layout=torch.sparse_csr)
tensor([0.6268, 0.6626, 0.9417, ..., 0.0422, 0.6229, 0.6981])
Matrix Type: SuiteSparse
Matrix: as-caida_G_070
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 78684
Density: 7.991130653390679e-05
Time: 22.265681505203247 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 78684, 78684, 78684]),
col_indices=tensor([ 106, 329, 1040, ..., 16263, 2242, 2242]),
values=tensor([1., 1., 1., ..., 3., 1., 1.]), size=(31379, 31379),
nnz=78684, layout=torch.sparse_csr)
tensor([0.6268, 0.6626, 0.9417, ..., 0.0422, 0.6229, 0.6981])
Matrix Type: SuiteSparse
Matrix: as-caida_G_070
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 78684
Density: 7.991130653390679e-05
Time: 22.265681505203247 seconds
[45.03, 39.14, 39.55, 40.05, 39.08, 39.41, 39.08, 39.05, 39.67, 38.97]
[108.92]
22.516743898391724
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 246006, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_070', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 78684, 'MATRIX_DENSITY': 7.991130653390679e-05, 'TIME_S': 22.265681505203247, 'TIME_S_1KI': 0.0905086928985604, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2452.5237454128264, 'W': 108.92}
[45.03, 39.14, 39.55, 40.05, 39.08, 39.41, 39.08, 39.05, 39.67, 38.97, 40.17, 39.34, 39.47, 38.92, 39.05, 39.51, 39.22, 38.87, 39.39, 38.99]
710.38
35.519
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 246006, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_070', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 78684, 'MATRIX_DENSITY': 7.991130653390679e-05, 'TIME_S': 22.265681505203247, 'TIME_S_1KI': 0.0905086928985604, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2452.5237454128264, 'W': 108.92, 'J_1KI': 9.969365565932645, 'W_1KI': 0.4427534287781599, 'W_D': 73.40100000000001, 'J_D': 1652.751518885851, 'W_D_1KI': 0.29837077144459895, 'J_D_1KI': 0.0012128597328707386}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_075.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 228835, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_075", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 97492, "MATRIX_DENSITY": 9.901267216465406e-05, "TIME_S": 20.558088302612305, "TIME_S_1KI": 0.08983804183194137, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2547.4816253471377, "W": 108.56, "J_1KI": 11.132395067831135, "W_1KI": 0.47440295409356087, "W_D": 72.922, "J_D": 1711.1961595759392, "W_D_1KI": 0.31866628793672297, "J_D_1KI": 0.0013925592148785063}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_075.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_075.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_075", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 97492, "MATRIX_DENSITY": 9.901267216465406e-05, "TIME_S": 0.09981632232666016}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 97491, 97491, 97492]),
col_indices=tensor([22754, 22754, 106, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=97492, layout=torch.sparse_csr)
tensor([0.7547, 0.7232, 0.9214, ..., 0.5453, 0.7893, 0.6720])
Matrix Type: SuiteSparse
Matrix: as-caida_G_075
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 97492
Density: 9.901267216465406e-05
Time: 0.09981632232666016 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '210386', '-m', 'matrices/as-caida_pruned/as-caida_G_075.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_075", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 97492, "MATRIX_DENSITY": 9.901267216465406e-05, "TIME_S": 19.3068745136261}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 97491, 97491, 97492]),
col_indices=tensor([22754, 22754, 106, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=97492, layout=torch.sparse_csr)
tensor([0.2501, 0.7708, 0.6340, ..., 0.1093, 0.9970, 0.8812])
Matrix Type: SuiteSparse
Matrix: as-caida_G_075
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 97492
Density: 9.901267216465406e-05
Time: 19.3068745136261 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '228835', '-m', 'matrices/as-caida_pruned/as-caida_G_075.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_075", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 97492, "MATRIX_DENSITY": 9.901267216465406e-05, "TIME_S": 20.558088302612305}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 97491, 97491, 97492]),
col_indices=tensor([22754, 22754, 106, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=97492, layout=torch.sparse_csr)
tensor([0.6824, 0.0363, 0.8858, ..., 0.8723, 0.9945, 0.3255])
Matrix Type: SuiteSparse
Matrix: as-caida_G_075
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 97492
Density: 9.901267216465406e-05
Time: 20.558088302612305 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 97491, 97491, 97492]),
col_indices=tensor([22754, 22754, 106, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=97492, layout=torch.sparse_csr)
tensor([0.6824, 0.0363, 0.8858, ..., 0.8723, 0.9945, 0.3255])
Matrix Type: SuiteSparse
Matrix: as-caida_G_075
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 97492
Density: 9.901267216465406e-05
Time: 20.558088302612305 seconds
[41.25, 39.57, 45.05, 39.28, 39.46, 39.08, 39.45, 39.03, 39.72, 39.35]
[108.56]
23.466116666793823
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 228835, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_075', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 97492, 'MATRIX_DENSITY': 9.901267216465406e-05, 'TIME_S': 20.558088302612305, 'TIME_S_1KI': 0.08983804183194137, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2547.4816253471377, 'W': 108.56}
[41.25, 39.57, 45.05, 39.28, 39.46, 39.08, 39.45, 39.03, 39.72, 39.35, 40.09, 39.03, 39.05, 38.95, 38.96, 38.88, 38.94, 38.94, 39.29, 39.47]
712.76
35.638
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 228835, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_075', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 97492, 'MATRIX_DENSITY': 9.901267216465406e-05, 'TIME_S': 20.558088302612305, 'TIME_S_1KI': 0.08983804183194137, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2547.4816253471377, 'W': 108.56, 'J_1KI': 11.132395067831135, 'W_1KI': 0.47440295409356087, 'W_D': 72.922, 'J_D': 1711.1961595759392, 'W_D_1KI': 0.31866628793672297, 'J_D_1KI': 0.0013925592148785063}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_080.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 231978, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_080", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 98112, "MATRIX_DENSITY": 9.964234287345156e-05, "TIME_S": 21.490782022476196, "TIME_S_1KI": 0.09264146609797566, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2630.7592759251593, "W": 108.57, "J_1KI": 11.340555035068666, "W_1KI": 0.4680185189974911, "W_D": 72.83899999999998, "J_D": 1764.9615446174141, "W_D_1KI": 0.31399098190345626, "J_D_1KI": 0.001353537757474658}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_080.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_080.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_080", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 98112, "MATRIX_DENSITY": 9.964234287345156e-05, "TIME_S": 0.12141084671020508}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 98111, 98111, 98112]),
col_indices=tensor([22754, 22754, 106, ..., 4133, 31329, 12170]),
values=tensor([1., 1., 1., ..., 3., 3., 1.]), size=(31379, 31379),
nnz=98112, layout=torch.sparse_csr)
tensor([0.5522, 0.8373, 0.3779, ..., 0.6358, 0.6326, 0.2271])
Matrix Type: SuiteSparse
Matrix: as-caida_G_080
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 98112
Density: 9.964234287345156e-05
Time: 0.12141084671020508 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '172966', '-m', 'matrices/as-caida_pruned/as-caida_G_080.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_080", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 98112, "MATRIX_DENSITY": 9.964234287345156e-05, "TIME_S": 15.657852172851562}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 98111, 98111, 98112]),
col_indices=tensor([22754, 22754, 106, ..., 4133, 31329, 12170]),
values=tensor([1., 1., 1., ..., 3., 3., 1.]), size=(31379, 31379),
nnz=98112, layout=torch.sparse_csr)
tensor([0.5579, 0.8962, 0.2299, ..., 0.1817, 0.7504, 0.1559])
Matrix Type: SuiteSparse
Matrix: as-caida_G_080
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 98112
Density: 9.964234287345156e-05
Time: 15.657852172851562 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '231978', '-m', 'matrices/as-caida_pruned/as-caida_G_080.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_080", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 98112, "MATRIX_DENSITY": 9.964234287345156e-05, "TIME_S": 21.490782022476196}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 98111, 98111, 98112]),
col_indices=tensor([22754, 22754, 106, ..., 4133, 31329, 12170]),
values=tensor([1., 1., 1., ..., 3., 3., 1.]), size=(31379, 31379),
nnz=98112, layout=torch.sparse_csr)
tensor([0.6706, 0.5375, 0.7370, ..., 0.5134, 0.1361, 0.9920])
Matrix Type: SuiteSparse
Matrix: as-caida_G_080
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 98112
Density: 9.964234287345156e-05
Time: 21.490782022476196 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 98111, 98111, 98112]),
col_indices=tensor([22754, 22754, 106, ..., 4133, 31329, 12170]),
values=tensor([1., 1., 1., ..., 3., 3., 1.]), size=(31379, 31379),
nnz=98112, layout=torch.sparse_csr)
tensor([0.6706, 0.5375, 0.7370, ..., 0.5134, 0.1361, 0.9920])
Matrix Type: SuiteSparse
Matrix: as-caida_G_080
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 98112
Density: 9.964234287345156e-05
Time: 21.490782022476196 seconds
[40.67, 39.43, 44.71, 39.49, 39.25, 39.48, 39.39, 39.44, 39.05, 39.19]
[108.57]
24.23099637031555
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 231978, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_080', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 98112, 'MATRIX_DENSITY': 9.964234287345156e-05, 'TIME_S': 21.490782022476196, 'TIME_S_1KI': 0.09264146609797566, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2630.7592759251593, 'W': 108.57}
[40.67, 39.43, 44.71, 39.49, 39.25, 39.48, 39.39, 39.44, 39.05, 39.19, 40.12, 39.45, 40.63, 39.92, 38.97, 38.98, 39.04, 38.9, 39.05, 38.9]
714.6200000000001
35.73100000000001
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 231978, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_080', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 98112, 'MATRIX_DENSITY': 9.964234287345156e-05, 'TIME_S': 21.490782022476196, 'TIME_S_1KI': 0.09264146609797566, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2630.7592759251593, 'W': 108.57, 'J_1KI': 11.340555035068666, 'W_1KI': 0.4680185189974911, 'W_D': 72.83899999999998, 'J_D': 1764.9615446174141, 'W_D_1KI': 0.31399098190345626, 'J_D_1KI': 0.001353537757474658}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_085.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 240739, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_085", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 99166, "MATRIX_DENSITY": 0.0001007127830784073, "TIME_S": 20.274125337600708, "TIME_S_1KI": 0.08421620650414227, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2593.0004249954227, "W": 110.26000000000002, "J_1KI": 10.771002724923767, "W_1KI": 0.45800638866157967, "W_D": 74.74925000000002, "J_D": 1757.8889626164441, "W_D_1KI": 0.31049912976293836, "J_D_1KI": 0.0012897749420033248}

85
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_085.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_085.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_085", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 99166, "MATRIX_DENSITY": 0.0001007127830784073, "TIME_S": 0.11759328842163086}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 99165, 99165, 99166]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=99166, layout=torch.sparse_csr)
tensor([0.7707, 0.4663, 0.0430, ..., 0.4487, 0.2025, 0.1450])
Matrix Type: SuiteSparse
Matrix: as-caida_G_085
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 99166
Density: 0.0001007127830784073
Time: 0.11759328842163086 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '178581', '-m', 'matrices/as-caida_pruned/as-caida_G_085.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_085", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 99166, "MATRIX_DENSITY": 0.0001007127830784073, "TIME_S": 15.577868700027466}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 99165, 99165, 99166]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=99166, layout=torch.sparse_csr)
tensor([0.9390, 0.2041, 0.9410, ..., 0.4323, 0.1487, 0.9127])
Matrix Type: SuiteSparse
Matrix: as-caida_G_085
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 99166
Density: 0.0001007127830784073
Time: 15.577868700027466 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '240739', '-m', 'matrices/as-caida_pruned/as-caida_G_085.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_085", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 99166, "MATRIX_DENSITY": 0.0001007127830784073, "TIME_S": 20.274125337600708}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 99165, 99165, 99166]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=99166, layout=torch.sparse_csr)
tensor([0.6109, 0.4136, 0.6946, ..., 0.9768, 0.6288, 0.8861])
Matrix Type: SuiteSparse
Matrix: as-caida_G_085
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 99166
Density: 0.0001007127830784073
Time: 20.274125337600708 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 99165, 99165, 99166]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=99166, layout=torch.sparse_csr)
tensor([0.6109, 0.4136, 0.6946, ..., 0.9768, 0.6288, 0.8861])
Matrix Type: SuiteSparse
Matrix: as-caida_G_085
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 99166
Density: 0.0001007127830784073
Time: 20.274125337600708 seconds
[39.65, 39.13, 39.02, 39.3, 39.12, 39.02, 39.24, 39.05, 39.96, 38.96]
[110.26]
23.51714515686035
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 240739, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_085', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 99166, 'MATRIX_DENSITY': 0.0001007127830784073, 'TIME_S': 20.274125337600708, 'TIME_S_1KI': 0.08421620650414227, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2593.0004249954227, 'W': 110.26000000000002}
[39.65, 39.13, 39.02, 39.3, 39.12, 39.02, 39.24, 39.05, 39.96, 38.96, 45.56, 39.23, 39.18, 39.55, 39.1, 39.1, 39.15, 39.32, 40.17, 38.98]
710.2149999999999
35.510749999999994
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 240739, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_085', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 99166, 'MATRIX_DENSITY': 0.0001007127830784073, 'TIME_S': 20.274125337600708, 'TIME_S_1KI': 0.08421620650414227, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2593.0004249954227, 'W': 110.26000000000002, 'J_1KI': 10.771002724923767, 'W_1KI': 0.45800638866157967, 'W_D': 74.74925000000002, 'J_D': 1757.8889626164441, 'W_D_1KI': 0.31049912976293836, 'J_D_1KI': 0.0012897749420033248}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_090.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 252609, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_090", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 100924, "MATRIX_DENSITY": 0.00010249820421722343, "TIME_S": 21.922589540481567, "TIME_S_1KI": 0.08678467331125006, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2682.2763265132903, "W": 110.86, "J_1KI": 10.618292802367653, "W_1KI": 0.43886005645087867, "W_D": 75.167, "J_D": 1818.6781944346428, "W_D_1KI": 0.2975626363272884, "J_D_1KI": 0.0011779573820698724}

89
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_090.output Normal file
View File

@ -0,0 +1,89 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_090.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_090", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 100924, "MATRIX_DENSITY": 0.00010249820421722343, "TIME_S": 0.09967470169067383}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 100923, 100923,
100924]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=100924, layout=torch.sparse_csr)
tensor([0.2862, 0.4561, 0.2506, ..., 0.8743, 0.1094, 0.9183])
Matrix Type: SuiteSparse
Matrix: as-caida_G_090
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 100924
Density: 0.00010249820421722343
Time: 0.09967470169067383 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '210685', '-m', 'matrices/as-caida_pruned/as-caida_G_090.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_090", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 100924, "MATRIX_DENSITY": 0.00010249820421722343, "TIME_S": 17.51469898223877}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 100923, 100923,
100924]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=100924, layout=torch.sparse_csr)
tensor([0.5217, 0.4687, 0.7241, ..., 0.7435, 0.7216, 0.9537])
Matrix Type: SuiteSparse
Matrix: as-caida_G_090
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 100924
Density: 0.00010249820421722343
Time: 17.51469898223877 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '252609', '-m', 'matrices/as-caida_pruned/as-caida_G_090.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_090", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 100924, "MATRIX_DENSITY": 0.00010249820421722343, "TIME_S": 21.922589540481567}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 100923, 100923,
100924]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=100924, layout=torch.sparse_csr)
tensor([0.1627, 0.9463, 0.5117, ..., 0.2113, 0.1004, 0.5160])
Matrix Type: SuiteSparse
Matrix: as-caida_G_090
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 100924
Density: 0.00010249820421722343
Time: 21.922589540481567 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 100923, 100923,
100924]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 31211, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=100924, layout=torch.sparse_csr)
tensor([0.1627, 0.9463, 0.5117, ..., 0.2113, 0.1004, 0.5160])
Matrix Type: SuiteSparse
Matrix: as-caida_G_090
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 100924
Density: 0.00010249820421722343
Time: 21.922589540481567 seconds
[40.69, 39.28, 39.34, 39.64, 40.84, 39.2, 39.95, 39.17, 39.41, 39.14]
[110.86]
24.195168018341064
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 252609, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_090', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 100924, 'MATRIX_DENSITY': 0.00010249820421722343, 'TIME_S': 21.922589540481567, 'TIME_S_1KI': 0.08678467331125006, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2682.2763265132903, 'W': 110.86}
[40.69, 39.28, 39.34, 39.64, 40.84, 39.2, 39.95, 39.17, 39.41, 39.14, 40.12, 39.33, 39.83, 39.09, 39.28, 41.26, 40.54, 39.1, 39.1, 39.05]
713.86
35.693
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 252609, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_090', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 100924, 'MATRIX_DENSITY': 0.00010249820421722343, 'TIME_S': 21.922589540481567, 'TIME_S_1KI': 0.08678467331125006, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2682.2763265132903, 'W': 110.86, 'J_1KI': 10.618292802367653, 'W_1KI': 0.43886005645087867, 'W_D': 75.167, 'J_D': 1818.6781944346428, 'W_D_1KI': 0.2975626363272884, 'J_D_1KI': 0.0011779573820698724}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_095.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 244511, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_095", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102290, "MATRIX_DENSITY": 0.00010388551097241275, "TIME_S": 21.708153247833252, "TIME_S_1KI": 0.08878190857602829, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2484.271987452507, "W": 110.49, "J_1KI": 10.160164522056295, "W_1KI": 0.45188151044329294, "W_D": 74.96725, "J_D": 1685.5737093976738, "W_D_1KI": 0.3066007255297308, "J_D_1KI": 0.0012539342832417796}

89
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_095.output Normal file
View File

@ -0,0 +1,89 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_095.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_095", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102290, "MATRIX_DENSITY": 0.00010388551097241275, "TIME_S": 0.0991964340209961}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102289, 102289,
102290]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 25970, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102290, layout=torch.sparse_csr)
tensor([0.8556, 0.9665, 0.3722, ..., 0.6225, 0.9450, 0.0036])
Matrix Type: SuiteSparse
Matrix: as-caida_G_095
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102290
Density: 0.00010388551097241275
Time: 0.0991964340209961 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '211701', '-m', 'matrices/as-caida_pruned/as-caida_G_095.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_095", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102290, "MATRIX_DENSITY": 0.00010388551097241275, "TIME_S": 18.18206286430359}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102289, 102289,
102290]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 25970, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102290, layout=torch.sparse_csr)
tensor([0.6510, 0.8867, 0.3325, ..., 0.9710, 0.7424, 0.5430])
Matrix Type: SuiteSparse
Matrix: as-caida_G_095
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102290
Density: 0.00010388551097241275
Time: 18.18206286430359 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '244511', '-m', 'matrices/as-caida_pruned/as-caida_G_095.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_095", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102290, "MATRIX_DENSITY": 0.00010388551097241275, "TIME_S": 21.708153247833252}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102289, 102289,
102290]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 25970, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102290, layout=torch.sparse_csr)
tensor([0.6961, 0.4709, 0.2900, ..., 0.8863, 0.2086, 0.2521])
Matrix Type: SuiteSparse
Matrix: as-caida_G_095
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102290
Density: 0.00010388551097241275
Time: 21.708153247833252 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102289, 102289,
102290]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 25970, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102290, layout=torch.sparse_csr)
tensor([0.6961, 0.4709, 0.2900, ..., 0.8863, 0.2086, 0.2521])
Matrix Type: SuiteSparse
Matrix: as-caida_G_095
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102290
Density: 0.00010388551097241275
Time: 21.708153247833252 seconds
[39.83, 39.41, 44.85, 39.09, 39.11, 39.54, 38.95, 38.94, 39.01, 39.32]
[110.49]
22.484134197235107
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 244511, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_095', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 102290, 'MATRIX_DENSITY': 0.00010388551097241275, 'TIME_S': 21.708153247833252, 'TIME_S_1KI': 0.08878190857602829, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2484.271987452507, 'W': 110.49}
[39.83, 39.41, 44.85, 39.09, 39.11, 39.54, 38.95, 38.94, 39.01, 39.32, 40.21, 39.31, 39.4, 38.98, 38.95, 38.93, 38.83, 38.78, 39.27, 38.85]
710.4549999999999
35.522749999999995
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 244511, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_095', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 102290, 'MATRIX_DENSITY': 0.00010388551097241275, 'TIME_S': 21.708153247833252, 'TIME_S_1KI': 0.08878190857602829, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2484.271987452507, 'W': 110.49, 'J_1KI': 10.160164522056295, 'W_1KI': 0.45188151044329294, 'W_D': 74.96725, 'J_D': 1685.5737093976738, 'W_D_1KI': 0.3066007255297308, 'J_D_1KI': 0.0012539342832417796}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_100.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 241218, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_100", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102888, "MATRIX_DENSITY": 0.00010449283852702711, "TIME_S": 20.706451177597046, "TIME_S_1KI": 0.08584123563580266, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2648.6161953759192, "W": 109.67, "J_1KI": 10.980176418741218, "W_1KI": 0.45465097961180345, "W_D": 73.90275, "J_D": 1784.8091595953106, "W_D_1KI": 0.30637328060095015, "J_D_1KI": 0.001270109529972681}

89
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_100.output Normal file
View File

@ -0,0 +1,89 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_100.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_100", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102888, "MATRIX_DENSITY": 0.00010449283852702711, "TIME_S": 0.10281991958618164}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102886, 102887,
102888]),
col_indices=tensor([ 106, 329, 1040, ..., 25970, 5128, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102888, layout=torch.sparse_csr)
tensor([0.4680, 0.0916, 0.1498, ..., 0.5442, 0.0289, 0.8913])
Matrix Type: SuiteSparse
Matrix: as-caida_G_100
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102888
Density: 0.00010449283852702711
Time: 0.10281991958618164 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '204240', '-m', 'matrices/as-caida_pruned/as-caida_G_100.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_100", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102888, "MATRIX_DENSITY": 0.00010449283852702711, "TIME_S": 17.780753135681152}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102886, 102887,
102888]),
col_indices=tensor([ 106, 329, 1040, ..., 25970, 5128, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102888, layout=torch.sparse_csr)
tensor([0.3783, 0.9598, 0.1242, ..., 0.4171, 0.3302, 0.6976])
Matrix Type: SuiteSparse
Matrix: as-caida_G_100
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102888
Density: 0.00010449283852702711
Time: 17.780753135681152 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '241218', '-m', 'matrices/as-caida_pruned/as-caida_G_100.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_100", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 102888, "MATRIX_DENSITY": 0.00010449283852702711, "TIME_S": 20.706451177597046}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102886, 102887,
102888]),
col_indices=tensor([ 106, 329, 1040, ..., 25970, 5128, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102888, layout=torch.sparse_csr)
tensor([0.7242, 0.2886, 0.2051, ..., 0.5324, 0.1969, 0.5524])
Matrix Type: SuiteSparse
Matrix: as-caida_G_100
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102888
Density: 0.00010449283852702711
Time: 20.706451177597046 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 102886, 102887,
102888]),
col_indices=tensor([ 106, 329, 1040, ..., 25970, 5128, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=102888, layout=torch.sparse_csr)
tensor([0.7242, 0.2886, 0.2051, ..., 0.5324, 0.1969, 0.5524])
Matrix Type: SuiteSparse
Matrix: as-caida_G_100
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 102888
Density: 0.00010449283852702711
Time: 20.706451177597046 seconds
[41.19, 39.42, 39.18, 39.49, 39.54, 39.15, 39.04, 39.04, 39.4, 38.9]
[109.67]
24.150781393051147
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 241218, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_100', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 102888, 'MATRIX_DENSITY': 0.00010449283852702711, 'TIME_S': 20.706451177597046, 'TIME_S_1KI': 0.08584123563580266, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2648.6161953759192, 'W': 109.67}
[41.19, 39.42, 39.18, 39.49, 39.54, 39.15, 39.04, 39.04, 39.4, 38.9, 39.85, 39.01, 40.47, 38.94, 39.0, 38.95, 39.45, 44.44, 41.2, 39.31]
715.3449999999999
35.76725
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 241218, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_100', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 102888, 'MATRIX_DENSITY': 0.00010449283852702711, 'TIME_S': 20.706451177597046, 'TIME_S_1KI': 0.08584123563580266, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2648.6161953759192, 'W': 109.67, 'J_1KI': 10.980176418741218, 'W_1KI': 0.45465097961180345, 'W_D': 73.90275, 'J_D': 1784.8091595953106, 'W_D_1KI': 0.30637328060095015, 'J_D_1KI': 0.001270109529972681}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_105.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 247484, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_105", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104726, "MATRIX_DENSITY": 0.00010635950749923647, "TIME_S": 20.726130485534668, "TIME_S_1KI": 0.08374735532614096, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2676.5612843990325, "W": 110.2, "J_1KI": 10.815088185090884, "W_1KI": 0.445281311115062, "W_D": 74.83, "J_D": 1817.4871226096152, "W_D_1KI": 0.30236298104119863, "J_D_1KI": 0.001221747591929978}

89
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_105.output Normal file
View File

@ -0,0 +1,89 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_105.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_105", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104726, "MATRIX_DENSITY": 0.00010635950749923647, "TIME_S": 0.10206460952758789}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104725, 104725,
104726]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104726, layout=torch.sparse_csr)
tensor([0.7941, 0.4112, 0.8474, ..., 0.8075, 0.4406, 0.7975])
Matrix Type: SuiteSparse
Matrix: as-caida_G_105
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104726
Density: 0.00010635950749923647
Time: 0.10206460952758789 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '205752', '-m', 'matrices/as-caida_pruned/as-caida_G_105.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_105", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104726, "MATRIX_DENSITY": 0.00010635950749923647, "TIME_S": 17.45883798599243}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104725, 104725,
104726]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104726, layout=torch.sparse_csr)
tensor([0.6509, 0.0632, 0.4912, ..., 0.8214, 0.1083, 0.9290])
Matrix Type: SuiteSparse
Matrix: as-caida_G_105
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104726
Density: 0.00010635950749923647
Time: 17.45883798599243 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '247484', '-m', 'matrices/as-caida_pruned/as-caida_G_105.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_105", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104726, "MATRIX_DENSITY": 0.00010635950749923647, "TIME_S": 20.726130485534668}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104725, 104725,
104726]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104726, layout=torch.sparse_csr)
tensor([0.8186, 0.7461, 0.3011, ..., 0.4943, 0.6376, 0.8092])
Matrix Type: SuiteSparse
Matrix: as-caida_G_105
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104726
Density: 0.00010635950749923647
Time: 20.726130485534668 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104725, 104725,
104726]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104726, layout=torch.sparse_csr)
tensor([0.8186, 0.7461, 0.3011, ..., 0.4943, 0.6376, 0.8092])
Matrix Type: SuiteSparse
Matrix: as-caida_G_105
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104726
Density: 0.00010635950749923647
Time: 20.726130485534668 seconds
[39.73, 39.49, 39.07, 39.13, 39.1, 39.5, 39.55, 39.36, 39.08, 38.97]
[110.2]
24.288214921951294
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 247484, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_105', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 104726, 'MATRIX_DENSITY': 0.00010635950749923647, 'TIME_S': 20.726130485534668, 'TIME_S_1KI': 0.08374735532614096, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2676.5612843990325, 'W': 110.2}
[39.73, 39.49, 39.07, 39.13, 39.1, 39.5, 39.55, 39.36, 39.08, 38.97, 39.81, 39.63, 39.02, 39.43, 39.27, 39.35, 39.73, 38.86, 39.07, 39.01]
707.4000000000001
35.370000000000005
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 247484, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_105', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 104726, 'MATRIX_DENSITY': 0.00010635950749923647, 'TIME_S': 20.726130485534668, 'TIME_S_1KI': 0.08374735532614096, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2676.5612843990325, 'W': 110.2, 'J_1KI': 10.815088185090884, 'W_1KI': 0.445281311115062, 'W_D': 74.83, 'J_D': 1817.4871226096152, 'W_D_1KI': 0.30236298104119863, 'J_D_1KI': 0.001221747591929978}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_110.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 225094, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_110", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104846, "MATRIX_DENSITY": 0.0001064813792493263, "TIME_S": 20.045295476913452, "TIME_S_1KI": 0.08905299775610835, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2434.8910455989835, "W": 110.26999999999998, "J_1KI": 10.817218786813436, "W_1KI": 0.48988422614552135, "W_D": 73.64949999999997, "J_D": 1626.2674169116015, "W_D_1KI": 0.3271944165548614, "J_D_1KI": 0.0014535901292564947}

89
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_110.output Normal file
View File

@ -0,0 +1,89 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_110.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_110", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104846, "MATRIX_DENSITY": 0.0001064813792493263, "TIME_S": 0.10591554641723633}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104844, 104844,
104846]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 2616, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104846, layout=torch.sparse_csr)
tensor([0.5817, 0.1599, 0.8730, ..., 0.2935, 0.5494, 0.8207])
Matrix Type: SuiteSparse
Matrix: as-caida_G_110
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104846
Density: 0.0001064813792493263
Time: 0.10591554641723633 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '198271', '-m', 'matrices/as-caida_pruned/as-caida_G_110.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_110", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104846, "MATRIX_DENSITY": 0.0001064813792493263, "TIME_S": 18.49750328063965}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104844, 104844,
104846]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 2616, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104846, layout=torch.sparse_csr)
tensor([0.4052, 0.0572, 0.8332, ..., 0.6912, 0.3718, 0.3744])
Matrix Type: SuiteSparse
Matrix: as-caida_G_110
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104846
Density: 0.0001064813792493263
Time: 18.49750328063965 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '225094', '-m', 'matrices/as-caida_pruned/as-caida_G_110.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_110", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 104846, "MATRIX_DENSITY": 0.0001064813792493263, "TIME_S": 20.045295476913452}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104844, 104844,
104846]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 2616, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104846, layout=torch.sparse_csr)
tensor([0.8447, 0.8415, 0.3714, ..., 0.3604, 0.8377, 0.5951])
Matrix Type: SuiteSparse
Matrix: as-caida_G_110
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104846
Density: 0.0001064813792493263
Time: 20.045295476913452 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 104844, 104844,
104846]),
col_indices=tensor([ 106, 329, 1040, ..., 882, 2616, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=104846, layout=torch.sparse_csr)
tensor([0.8447, 0.8415, 0.3714, ..., 0.3604, 0.8377, 0.5951])
Matrix Type: SuiteSparse
Matrix: as-caida_G_110
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 104846
Density: 0.0001064813792493263
Time: 20.045295476913452 seconds
[40.62, 39.64, 39.23, 39.67, 39.73, 39.09, 39.6, 39.58, 39.51, 39.98]
[110.27]
22.08117389678955
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 225094, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_110', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 104846, 'MATRIX_DENSITY': 0.0001064813792493263, 'TIME_S': 20.045295476913452, 'TIME_S_1KI': 0.08905299775610835, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2434.8910455989835, 'W': 110.26999999999998}
[40.62, 39.64, 39.23, 39.67, 39.73, 39.09, 39.6, 39.58, 39.51, 39.98, 39.85, 39.46, 39.55, 39.26, 45.01, 39.42, 39.52, 49.35, 45.03, 39.07]
732.4100000000001
36.62050000000001
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 225094, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_110', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 104846, 'MATRIX_DENSITY': 0.0001064813792493263, 'TIME_S': 20.045295476913452, 'TIME_S_1KI': 0.08905299775610835, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2434.8910455989835, 'W': 110.26999999999998, 'J_1KI': 10.817218786813436, 'W_1KI': 0.48988422614552135, 'W_D': 73.64949999999997, 'J_D': 1626.2674169116015, 'W_D_1KI': 0.3271944165548614, 'J_D_1KI': 0.0014535901292564947}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_115.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 234073, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_115", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106312, "MATRIX_DENSITY": 0.00010797024579625715, "TIME_S": 21.110437870025635, "TIME_S_1KI": 0.09018741106417927, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2580.228788130283, "W": 109.77, "J_1KI": 11.023179897426372, "W_1KI": 0.4689562657803335, "W_D": 71.584, "J_D": 1682.6373104629517, "W_D_1KI": 0.3058191248029461, "J_D_1KI": 0.0013065117497658683}

89
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_115.output Normal file
View File

@ -0,0 +1,89 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_115.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_115", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106312, "MATRIX_DENSITY": 0.00010797024579625715, "TIME_S": 0.10738468170166016}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106311, 106311,
106312]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106312, layout=torch.sparse_csr)
tensor([0.0691, 0.6464, 0.3795, ..., 0.4403, 0.3685, 0.1679])
Matrix Type: SuiteSparse
Matrix: as-caida_G_115
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106312
Density: 0.00010797024579625715
Time: 0.10738468170166016 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '195558', '-m', 'matrices/as-caida_pruned/as-caida_G_115.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_115", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106312, "MATRIX_DENSITY": 0.00010797024579625715, "TIME_S": 17.54453992843628}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106311, 106311,
106312]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106312, layout=torch.sparse_csr)
tensor([0.3421, 0.1187, 0.2163, ..., 0.8976, 0.3011, 0.1448])
Matrix Type: SuiteSparse
Matrix: as-caida_G_115
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106312
Density: 0.00010797024579625715
Time: 17.54453992843628 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '234073', '-m', 'matrices/as-caida_pruned/as-caida_G_115.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_115", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106312, "MATRIX_DENSITY": 0.00010797024579625715, "TIME_S": 21.110437870025635}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106311, 106311,
106312]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106312, layout=torch.sparse_csr)
tensor([0.9599, 0.3879, 0.7672, ..., 0.8941, 0.8822, 0.9833])
Matrix Type: SuiteSparse
Matrix: as-caida_G_115
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106312
Density: 0.00010797024579625715
Time: 21.110437870025635 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106311, 106311,
106312]),
col_indices=tensor([ 106, 329, 1040, ..., 160, 882, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106312, layout=torch.sparse_csr)
tensor([0.9599, 0.3879, 0.7672, ..., 0.8941, 0.8822, 0.9833])
Matrix Type: SuiteSparse
Matrix: as-caida_G_115
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106312
Density: 0.00010797024579625715
Time: 21.110437870025635 seconds
[39.49, 39.1, 39.1, 39.5, 39.4, 39.39, 39.24, 38.91, 38.94, 39.59]
[109.77]
23.505773782730103
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 234073, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_115', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 106312, 'MATRIX_DENSITY': 0.00010797024579625715, 'TIME_S': 21.110437870025635, 'TIME_S_1KI': 0.09018741106417927, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2580.228788130283, 'W': 109.77}
[39.49, 39.1, 39.1, 39.5, 39.4, 39.39, 39.24, 38.91, 38.94, 39.59, 39.66, 38.99, 39.64, 39.13, 38.92, 39.41, 39.02, 56.35, 67.32, 63.98]
763.72
38.186
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 234073, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_115', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 106312, 'MATRIX_DENSITY': 0.00010797024579625715, 'TIME_S': 21.110437870025635, 'TIME_S_1KI': 0.09018741106417927, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2580.228788130283, 'W': 109.77, 'J_1KI': 11.023179897426372, 'W_1KI': 0.4689562657803335, 'W_D': 71.584, 'J_D': 1682.6373104629517, 'W_D_1KI': 0.3058191248029461, 'J_D_1KI': 0.0013065117497658683}

1
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_120.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "CORES": 16, "ITERATIONS": 246921, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_120", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106510, "MATRIX_DENSITY": 0.0001081713341839054, "TIME_S": 21.179029941558838, "TIME_S_1KI": 0.08577249379987462, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2789.262641429901, "W": 109.9, "J_1KI": 11.296174247754955, "W_1KI": 0.4450816252971599, "W_D": 74.2125, "J_D": 1883.513683140278, "W_D_1KI": 0.3005515934246176, "J_D_1KI": 0.001217197376588535}

89
pytorch/output_as-caida_maxcore/epyc_7313p_max_csr_20_10_10_as-caida_G_120.output Normal file
View File

@ -0,0 +1,89 @@
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_120.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_120", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106510, "MATRIX_DENSITY": 0.0001081713341839054, "TIME_S": 0.13150310516357422}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106509, 106509,
106510]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106510, layout=torch.sparse_csr)
tensor([0.0187, 0.2029, 0.6613, ..., 0.2021, 0.3325, 0.5912])
Matrix Type: SuiteSparse
Matrix: as-caida_G_120
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106510
Density: 0.0001081713341839054
Time: 0.13150310516357422 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '159692', '-m', 'matrices/as-caida_pruned/as-caida_G_120.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_120", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106510, "MATRIX_DENSITY": 0.0001081713341839054, "TIME_S": 13.581367492675781}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106509, 106509,
106510]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106510, layout=torch.sparse_csr)
tensor([0.0016, 0.7280, 0.4677, ..., 0.1439, 0.6254, 0.9086])
Matrix Type: SuiteSparse
Matrix: as-caida_G_120
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106510
Density: 0.0001081713341839054
Time: 13.581367492675781 seconds
['apptainer', 'run', 'pytorch-epyc_7313p.sif', 'python3', 'spmv.py', 'suitesparse', 'csr', '246921', '-m', 'matrices/as-caida_pruned/as-caida_G_120.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_120", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 106510, "MATRIX_DENSITY": 0.0001081713341839054, "TIME_S": 21.179029941558838}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106509, 106509,
106510]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106510, layout=torch.sparse_csr)
tensor([0.4228, 0.0866, 0.5391, ..., 0.1734, 0.5961, 0.3644])
Matrix Type: SuiteSparse
Matrix: as-caida_G_120
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106510
Density: 0.0001081713341839054
Time: 21.179029941558838 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 106509, 106509,
106510]),
col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]),
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=106510, layout=torch.sparse_csr)
tensor([0.4228, 0.0866, 0.5391, ..., 0.1734, 0.5961, 0.3644])
Matrix Type: SuiteSparse
Matrix: as-caida_G_120
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 106510
Density: 0.0001081713341839054
Time: 21.179029941558838 seconds
[40.0, 44.52, 39.99, 39.55, 39.78, 39.63, 39.26, 39.08, 39.04, 39.39]
[109.9]
25.380005836486816
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 246921, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_120', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 106510, 'MATRIX_DENSITY': 0.0001081713341839054, 'TIME_S': 21.179029941558838, 'TIME_S_1KI': 0.08577249379987462, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2789.262641429901, 'W': 109.9}
[40.0, 44.52, 39.99, 39.55, 39.78, 39.63, 39.26, 39.08, 39.04, 39.39, 39.72, 39.14, 39.01, 39.02, 39.02, 39.73, 39.52, 39.34, 38.96, 39.21]
713.75
35.6875
{'CPU': 'Epyc 7313P', 'CORES': 16, 'ITERATIONS': 246921, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_120', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 106510, 'MATRIX_DENSITY': 0.0001081713341839054, 'TIME_S': 21.179029941558838, 'TIME_S_1KI': 0.08577249379987462, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2789.262641429901, 'W': 109.9, 'J_1KI': 11.296174247754955, 'W_1KI': 0.4450816252971599, 'W_D': 74.2125, 'J_D': 1883.513683140278, 'W_D_1KI': 0.3005515934246176, 'J_D_1KI': 0.001217197376588535}

1
pytorch/output_as-caida_maxcore/xeon_4216_max_csr_20_10_10_as-caida_G_005.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "CORES": 16, "ITERATIONS": 228681, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 20.71348738670349, "TIME_S_1KI": 0.09057808644663742, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 1985.7099216771128, "W": 81.43, "J_1KI": 8.683318341607361, "W_1KI": 0.35608555148875515, "W_D": 64.78550000000001, "J_D": 1579.8257415057424, "W_D_1KI": 0.2833007552004758, "J_D_1KI": 0.0012388469317541721}

85
pytorch/output_as-caida_maxcore/xeon_4216_max_csr_20_10_10_as-caida_G_005.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-xeon_4216.sif', 'numactl', '--cpunodebind=0', '--membind=0', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_005.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 0.10603070259094238}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.7270, 0.4992, 0.4890, ..., 0.7263, 0.9568, 0.8257])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 0.10603070259094238 seconds
['apptainer', 'run', 'pytorch-xeon_4216.sif', 'numactl', '--cpunodebind=0', '--membind=0', 'python3', 'spmv.py', 'suitesparse', 'csr', '198055', '-m', 'matrices/as-caida_pruned/as-caida_G_005.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 18.187560081481934}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.8953, 0.8779, 0.8892, ..., 0.4124, 0.7171, 0.6644])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 18.187560081481934 seconds
['apptainer', 'run', 'pytorch-xeon_4216.sif', 'numactl', '--cpunodebind=0', '--membind=0', 'python3', 'spmv.py', 'suitesparse', 'csr', '228681', '-m', 'matrices/as-caida_pruned/as-caida_G_005.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_005", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 70026, "MATRIX_DENSITY": 7.111825976492498e-05, "TIME_S": 20.71348738670349}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.1790, 0.8570, 0.6154, ..., 0.3543, 0.1752, 0.3411])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 20.71348738670349 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 63, 63, ..., 70025, 70025, 70026]),
col_indices=tensor([ 111, 761, 822, ..., 978, 978, 12170]),
values=tensor([4., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=70026, layout=torch.sparse_csr)
tensor([0.1790, 0.8570, 0.6154, ..., 0.3543, 0.1752, 0.3411])
Matrix Type: SuiteSparse
Matrix: as-caida_G_005
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 70026
Density: 7.111825976492498e-05
Time: 20.71348738670349 seconds
[18.75, 18.36, 18.44, 17.9, 21.42, 17.95, 18.08, 18.0, 18.15, 17.98]
[81.43]
24.385483503341675
{'CPU': 'Xeon 4216', 'CORES': 16, 'ITERATIONS': 228681, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_005', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 70026, 'MATRIX_DENSITY': 7.111825976492498e-05, 'TIME_S': 20.71348738670349, 'TIME_S_1KI': 0.09057808644663742, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 1985.7099216771128, 'W': 81.43}
[18.75, 18.36, 18.44, 17.9, 21.42, 17.95, 18.08, 18.0, 18.15, 17.98, 18.41, 17.76, 18.35, 21.01, 17.94, 18.41, 18.12, 18.12, 18.23, 18.16]
332.89
16.6445
{'CPU': 'Xeon 4216', 'CORES': 16, 'ITERATIONS': 228681, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_005', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 70026, 'MATRIX_DENSITY': 7.111825976492498e-05, 'TIME_S': 20.71348738670349, 'TIME_S_1KI': 0.09057808644663742, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 1985.7099216771128, 'W': 81.43, 'J_1KI': 8.683318341607361, 'W_1KI': 0.35608555148875515, 'W_D': 64.78550000000001, 'J_D': 1579.8257415057424, 'W_D_1KI': 0.2833007552004758, 'J_D_1KI': 0.0012388469317541721}

1
pytorch/output_as-caida_maxcore/xeon_4216_max_csr_20_10_10_as-caida_G_010.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "CORES": 16, "ITERATIONS": 219634, "MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 21.155344486236572, "TIME_S_1KI": 0.09632089970695144, "BASELINE_TIME_S": 10, "BASELINE_DELAY_S": 10, "J": 2007.079451637268, "W": 81.32, "J_1KI": 9.138291210091644, "W_1KI": 0.3702523288744001, "W_D": 65.02274999999999, "J_D": 1604.842909664869, "W_D_1KI": 0.2960504748809382, "J_D_1KI": 0.0013479264361662504}

85
pytorch/output_as-caida_maxcore/xeon_4216_max_csr_20_10_10_as-caida_G_010.output Normal file
View File

@ -0,0 +1,85 @@
['apptainer', 'run', 'pytorch-xeon_4216.sif', 'numactl', '--cpunodebind=0', '--membind=0', 'python3', 'spmv.py', 'suitesparse', 'csr', '1000', '-m', 'matrices/as-caida_pruned/as-caida_G_010.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 0.10599613189697266}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.8277, 0.3875, 0.9175, ..., 0.7822, 0.3026, 0.5609])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 0.10599613189697266 seconds
['apptainer', 'run', 'pytorch-xeon_4216.sif', 'numactl', '--cpunodebind=0', '--membind=0', 'python3', 'spmv.py', 'suitesparse', 'csr', '198120', '-m', 'matrices/as-caida_pruned/as-caida_G_010.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 18.942925214767456}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.9981, 0.2399, 0.7684, ..., 0.8350, 0.7877, 0.0239])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 18.942925214767456 seconds
['apptainer', 'run', 'pytorch-xeon_4216.sif', 'numactl', '--cpunodebind=0', '--membind=0', 'python3', 'spmv.py', 'suitesparse', 'csr', '219634', '-m', 'matrices/as-caida_pruned/as-caida_G_010.mtx']
{"MATRIX_TYPE": "SuiteSparse", "MATRIX_FILE": "as-caida_G_010", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [31379, 31379], "MATRIX_ROWS": 31379, "MATRIX_SIZE": 984641641, "MATRIX_NNZ": 74994, "MATRIX_DENSITY": 7.616375021864427e-05, "TIME_S": 21.155344486236572}
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.3439, 0.2176, 0.6511, ..., 0.3157, 0.5036, 0.0852])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 21.155344486236572 seconds
/nfshomes/vut/ampere_research/pytorch/spmv.py:75: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 28, 28, ..., 74993, 74993, 74994]),
col_indices=tensor([ 1040, 2020, 2054, ..., 160, 160, 12170]),
values=tensor([1., 3., 3., ..., 1., 1., 1.]), size=(31379, 31379),
nnz=74994, layout=torch.sparse_csr)
tensor([0.3439, 0.2176, 0.6511, ..., 0.3157, 0.5036, 0.0852])
Matrix Type: SuiteSparse
Matrix: as-caida_G_010
Matrix Format: csr
Shape: torch.Size([31379, 31379])
Rows: 31379
Size: 984641641
NNZ: 74994
Density: 7.616375021864427e-05
Time: 21.155344486236572 seconds
[18.51, 18.19, 18.29, 17.98, 18.25, 17.99, 18.0, 18.01, 18.19, 18.67]
[81.32]
24.681252479553223
{'CPU': 'Xeon 4216', 'CORES': 16, 'ITERATIONS': 219634, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_010', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 74994, 'MATRIX_DENSITY': 7.616375021864427e-05, 'TIME_S': 21.155344486236572, 'TIME_S_1KI': 0.09632089970695144, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2007.079451637268, 'W': 81.32}
[18.51, 18.19, 18.29, 17.98, 18.25, 17.99, 18.0, 18.01, 18.19, 18.67, 18.4, 17.89, 18.25, 18.19, 17.92, 18.04, 17.99, 17.96, 18.08, 17.87]
325.94500000000005
16.297250000000002
{'CPU': 'Xeon 4216', 'CORES': 16, 'ITERATIONS': 219634, 'MATRIX_TYPE': 'SuiteSparse', 'MATRIX_FILE': 'as-caida_G_010', 'MATRIX_FORMAT': 'csr', 'MATRIX_SHAPE': [31379, 31379], 'MATRIX_ROWS': 31379, 'MATRIX_SIZE': 984641641, 'MATRIX_NNZ': 74994, 'MATRIX_DENSITY': 7.616375021864427e-05, 'TIME_S': 21.155344486236572, 'TIME_S_1KI': 0.09632089970695144, 'BASELINE_TIME_S': 10, 'BASELINE_DELAY_S': 10, 'J': 2007.079451637268, 'W': 81.32, 'J_1KI': 9.138291210091644, 'W_1KI': 0.3702523288744001, 'W_D': 65.02274999999999, 'J_D': 1604.842909664869, 'W_D_1KI': 0.2960504748809382, 'J_D_1KI': 0.0013479264361662504}

Some files were not shown because too many files have changed in this diff Show More