164 lines
9.8 KiB
Plaintext
164 lines
9.8 KiB
Plaintext
srun: Job time limit was unset; set to partition default of 60 minutes
|
|
srun: ################################################################################
|
|
srun: # Please note that the oasis compute nodes have aarch64 architecture CPUs. #
|
|
srun: # All submission nodes and all other compute nodes have x86_64 architecture #
|
|
srun: # CPUs. Programs, environments, or other software that was built on x86_64 #
|
|
srun: # nodes may need to be rebuilt to properly execute on these nodes. #
|
|
srun: ################################################################################
|
|
srun: job 3394979 queued and waiting for resources
|
|
srun: job 3394979 has been allocated resources
|
|
/nfshomes/vut/ampere_research/pytorch/spmv.py:20: 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.)
|
|
).to_sparse_csr().type(torch.float)
|
|
tensor(crow_indices=tensor([ 0, 29, 124, ..., 549200, 549200,
|
|
549202]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 82142, 81974, 82136]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(82144, 82144),
|
|
nnz=549202, layout=torch.sparse_csr)
|
|
tensor([0.4201, 0.7748, 0.6565, ..., 0.0517, 0.6958, 0.5341])
|
|
Matrix: soc-sign-Slashdot090221
|
|
Shape: torch.Size([82144, 82144])
|
|
NNZ: 549202
|
|
Density: 8.13917555860553e-05
|
|
Time: 27.35153603553772 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/soc-sign-Slashdot090221.mtx 1000':
|
|
|
|
58.57 msec task-clock:u # 0.002 CPUs utilized
|
|
0 context-switches:u # 0.000 /sec
|
|
0 cpu-migrations:u # 0.000 /sec
|
|
3,259 page-faults:u # 55.640 K/sec
|
|
74,509,373 cycles:u # 1.272 GHz (58.00%)
|
|
88,672,751 instructions:u # 1.19 insn per cycle (90.97%)
|
|
<not supported> branches:u
|
|
361,568 branch-misses:u
|
|
31,594,797 L1-dcache-loads:u # 539.410 M/sec
|
|
460,467 L1-dcache-load-misses:u # 1.46% of all L1-dcache accesses
|
|
<not supported> LLC-loads:u
|
|
<not supported> LLC-load-misses:u
|
|
30,148,838 L1-icache-loads:u # 514.724 M/sec
|
|
282,768 L1-icache-load-misses:u # 0.94% of all L1-icache accesses
|
|
19,757,856 dTLB-loads:u # 337.321 M/sec (11.69%)
|
|
<not counted> dTLB-load-misses:u (0.00%)
|
|
<not counted> iTLB-loads:u (0.00%)
|
|
<not counted> iTLB-load-misses:u (0.00%)
|
|
|
|
31.087250856 seconds time elapsed
|
|
|
|
142.716222000 seconds user
|
|
2102.420776000 seconds sys
|
|
|
|
|
|
|
|
/nfshomes/vut/ampere_research/pytorch/spmv.py:20: 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.)
|
|
).to_sparse_csr().type(torch.float)
|
|
tensor(crow_indices=tensor([ 0, 29, 124, ..., 549200, 549200,
|
|
549202]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 82142, 81974, 82136]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(82144, 82144),
|
|
nnz=549202, layout=torch.sparse_csr)
|
|
tensor([0.7637, 0.5328, 0.8286, ..., 0.7084, 0.8903, 0.1707])
|
|
Matrix: soc-sign-Slashdot090221
|
|
Shape: torch.Size([82144, 82144])
|
|
NNZ: 549202
|
|
Density: 8.13917555860553e-05
|
|
Time: 17.188836336135864 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/soc-sign-Slashdot090221.mtx 1000':
|
|
|
|
342,121 BR_MIS_PRED_RETIRED:u # 0.0 per branch branch_misprediction_ratio
|
|
20,436,338 BR_RETIRED:u
|
|
|
|
20.753346873 seconds time elapsed
|
|
|
|
98.605331000 seconds user
|
|
1332.291974000 seconds sys
|
|
|
|
|
|
|
|
/nfshomes/vut/ampere_research/pytorch/spmv.py:20: 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.)
|
|
).to_sparse_csr().type(torch.float)
|
|
tensor(crow_indices=tensor([ 0, 29, 124, ..., 549200, 549200,
|
|
549202]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 82142, 81974, 82136]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(82144, 82144),
|
|
nnz=549202, layout=torch.sparse_csr)
|
|
tensor([0.9017, 0.8505, 0.0023, ..., 0.4182, 0.6895, 0.5023])
|
|
Matrix: soc-sign-Slashdot090221
|
|
Shape: torch.Size([82144, 82144])
|
|
NNZ: 549202
|
|
Density: 8.13917555860553e-05
|
|
Time: 16.22375249862671 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/soc-sign-Slashdot090221.mtx 1000':
|
|
|
|
27,189,335 L1I_TLB:u # 0.0 per TLB access itlb_walk_ratio
|
|
6,437 ITLB_WALK:u
|
|
18,156 DTLB_WALK:u # 0.0 per TLB access dtlb_walk_ratio
|
|
36,676,625 L1D_TLB:u
|
|
|
|
19.748749363 seconds time elapsed
|
|
|
|
103.049578000 seconds user
|
|
1249.814927000 seconds sys
|
|
|
|
|
|
|
|
/nfshomes/vut/ampere_research/pytorch/spmv.py:20: 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.)
|
|
).to_sparse_csr().type(torch.float)
|
|
tensor(crow_indices=tensor([ 0, 29, 124, ..., 549200, 549200,
|
|
549202]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 82142, 81974, 82136]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(82144, 82144),
|
|
nnz=549202, layout=torch.sparse_csr)
|
|
tensor([0.4805, 0.2325, 0.2103, ..., 0.1710, 0.7638, 0.9368])
|
|
Matrix: soc-sign-Slashdot090221
|
|
Shape: torch.Size([82144, 82144])
|
|
NNZ: 549202
|
|
Density: 8.13917555860553e-05
|
|
Time: 15.453373908996582 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/soc-sign-Slashdot090221.mtx 1000':
|
|
|
|
30,721,032 L1I_CACHE:u # 0.0 per cache access l1i_cache_miss_ratio
|
|
302,777 L1I_CACHE_REFILL:u
|
|
469,833 L1D_CACHE_REFILL:u # 0.0 per cache access l1d_cache_miss_ratio
|
|
32,109,077 L1D_CACHE:u
|
|
|
|
19.090250444 seconds time elapsed
|
|
|
|
94.904880000 seconds user
|
|
1195.102767000 seconds sys
|
|
|
|
|
|
|
|
/nfshomes/vut/ampere_research/pytorch/spmv.py:20: 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.)
|
|
).to_sparse_csr().type(torch.float)
|
|
tensor(crow_indices=tensor([ 0, 29, 124, ..., 549200, 549200,
|
|
549202]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 82142, 81974, 82136]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(82144, 82144),
|
|
nnz=549202, layout=torch.sparse_csr)
|
|
tensor([0.8430, 0.9439, 0.4260, ..., 0.8172, 0.4243, 0.3834])
|
|
Matrix: soc-sign-Slashdot090221
|
|
Shape: torch.Size([82144, 82144])
|
|
NNZ: 549202
|
|
Density: 8.13917555860553e-05
|
|
Time: 29.316507816314697 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/soc-sign-Slashdot090221.mtx 1000':
|
|
|
|
551,850 LL_CACHE_MISS_RD:u # 1.0 per cache access ll_cache_read_miss_ratio
|
|
565,355 LL_CACHE_RD:u
|
|
200,417 L2D_TLB:u # 0.1 per TLB access l2_tlb_miss_ratio
|
|
25,536 L2D_TLB_REFILL:u
|
|
304,133 L2D_CACHE_REFILL:u # 0.2 per cache access l2_cache_miss_ratio
|
|
1,801,849 L2D_CACHE:u
|
|
|
|
32.859276963 seconds time elapsed
|
|
|
|
148.969816000 seconds user
|
|
2252.321936000 seconds sys
|
|
|
|
|
|
|