164 lines
9.7 KiB
Plaintext
164 lines
9.7 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 3394984 queued and waiting for resources
|
|
srun: job 3394984 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, 3, 8, ..., 125742, 125747,
|
|
125750]),
|
|
col_indices=tensor([ 25, 56, 662, ..., 21738, 22279, 23882]),
|
|
values=tensor([17171., 37318., 5284., ..., 25993., 24918., 803.]),
|
|
size=(25181, 25181), nnz=125750, layout=torch.sparse_csr)
|
|
tensor([0.5906, 0.9651, 0.2033, ..., 0.2175, 0.4484, 0.0412])
|
|
Matrix: ri2010
|
|
Shape: torch.Size([25181, 25181])
|
|
NNZ: 125750
|
|
Density: 0.00019831796057928155
|
|
Time: 3.107008934020996 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/ri2010.mtx 1000':
|
|
|
|
64.49 msec task-clock:u # 0.010 CPUs utilized
|
|
0 context-switches:u # 0.000 /sec
|
|
0 cpu-migrations:u # 0.000 /sec
|
|
3,473 page-faults:u # 53.852 K/sec
|
|
42,783,607 cycles:u # 0.663 GHz (37.27%)
|
|
84,598,454 instructions:u # 1.98 insn per cycle (73.53%)
|
|
<not supported> branches:u
|
|
353,558 branch-misses:u (89.57%)
|
|
33,192,964 L1-dcache-loads:u # 514.689 M/sec
|
|
466,217 L1-dcache-load-misses:u # 1.40% of all L1-dcache accesses
|
|
<not supported> LLC-loads:u
|
|
<not supported> LLC-load-misses:u
|
|
31,727,502 L1-icache-loads:u # 491.965 M/sec
|
|
292,570 L1-icache-load-misses:u # 0.92% of all L1-icache accesses
|
|
38,623,737 dTLB-loads:u # 598.898 M/sec (34.88%)
|
|
124,174 dTLB-load-misses:u # 0.32% of all dTLB cache accesses (14.74%)
|
|
<not counted> iTLB-loads:u (0.00%)
|
|
<not counted> iTLB-load-misses:u (0.00%)
|
|
|
|
6.612563197 seconds time elapsed
|
|
|
|
18.114584000 seconds user
|
|
29.808542000 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, 3, 8, ..., 125742, 125747,
|
|
125750]),
|
|
col_indices=tensor([ 25, 56, 662, ..., 21738, 22279, 23882]),
|
|
values=tensor([17171., 37318., 5284., ..., 25993., 24918., 803.]),
|
|
size=(25181, 25181), nnz=125750, layout=torch.sparse_csr)
|
|
tensor([0.6092, 0.5511, 0.6052, ..., 0.8002, 0.0295, 0.2972])
|
|
Matrix: ri2010
|
|
Shape: torch.Size([25181, 25181])
|
|
NNZ: 125750
|
|
Density: 0.00019831796057928155
|
|
Time: 2.9385879039764404 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/ri2010.mtx 1000':
|
|
|
|
331,326 BR_MIS_PRED_RETIRED:u # 0.0 per branch branch_misprediction_ratio
|
|
20,438,455 BR_RETIRED:u
|
|
|
|
6.446731410 seconds time elapsed
|
|
|
|
17.939571000 seconds user
|
|
33.272929000 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, 3, 8, ..., 125742, 125747,
|
|
125750]),
|
|
col_indices=tensor([ 25, 56, 662, ..., 21738, 22279, 23882]),
|
|
values=tensor([17171., 37318., 5284., ..., 25993., 24918., 803.]),
|
|
size=(25181, 25181), nnz=125750, layout=torch.sparse_csr)
|
|
tensor([0.3348, 0.2974, 0.2569, ..., 0.2397, 0.1965, 0.5651])
|
|
Matrix: ri2010
|
|
Shape: torch.Size([25181, 25181])
|
|
NNZ: 125750
|
|
Density: 0.00019831796057928155
|
|
Time: 2.972891330718994 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/ri2010.mtx 1000':
|
|
|
|
26,869,742 L1I_TLB:u # 0.0 per TLB access itlb_walk_ratio
|
|
6,302 ITLB_WALK:u
|
|
14,926 DTLB_WALK:u # 0.0 per TLB access dtlb_walk_ratio
|
|
36,876,841 L1D_TLB:u
|
|
|
|
6.376775396 seconds time elapsed
|
|
|
|
17.836418000 seconds user
|
|
29.830135000 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, 3, 8, ..., 125742, 125747,
|
|
125750]),
|
|
col_indices=tensor([ 25, 56, 662, ..., 21738, 22279, 23882]),
|
|
values=tensor([17171., 37318., 5284., ..., 25993., 24918., 803.]),
|
|
size=(25181, 25181), nnz=125750, layout=torch.sparse_csr)
|
|
tensor([0.7889, 0.7395, 0.6553, ..., 0.3938, 0.2478, 0.7923])
|
|
Matrix: ri2010
|
|
Shape: torch.Size([25181, 25181])
|
|
NNZ: 125750
|
|
Density: 0.00019831796057928155
|
|
Time: 2.9658284187316895 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/ri2010.mtx 1000':
|
|
|
|
31,664,385 L1I_CACHE:u # 0.0 per cache access l1i_cache_miss_ratio
|
|
301,678 L1I_CACHE_REFILL:u
|
|
493,536 L1D_CACHE_REFILL:u # 0.0 per cache access l1d_cache_miss_ratio
|
|
33,219,437 L1D_CACHE:u
|
|
|
|
6.559158078 seconds time elapsed
|
|
|
|
19.008146000 seconds user
|
|
38.233666000 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, 3, 8, ..., 125742, 125747,
|
|
125750]),
|
|
col_indices=tensor([ 25, 56, 662, ..., 21738, 22279, 23882]),
|
|
values=tensor([17171., 37318., 5284., ..., 25993., 24918., 803.]),
|
|
size=(25181, 25181), nnz=125750, layout=torch.sparse_csr)
|
|
tensor([0.1256, 0.1417, 0.9800, ..., 0.2509, 0.8121, 0.6210])
|
|
Matrix: ri2010
|
|
Shape: torch.Size([25181, 25181])
|
|
NNZ: 125750
|
|
Density: 0.00019831796057928155
|
|
Time: 2.9228267669677734 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/ri2010.mtx 1000':
|
|
|
|
552,180 LL_CACHE_MISS_RD:u # 1.0 per cache access ll_cache_read_miss_ratio
|
|
564,990 LL_CACHE_RD:u
|
|
167,824 L2D_TLB:u # 0.1 per TLB access l2_tlb_miss_ratio
|
|
19,594 L2D_TLB_REFILL:u
|
|
304,114 L2D_CACHE_REFILL:u # 0.2 per cache access l2_cache_miss_ratio
|
|
1,716,370 L2D_CACHE:u
|
|
|
|
6.135787277 seconds time elapsed
|
|
|
|
18.029630000 seconds user
|
|
28.723217000 seconds sys
|
|
|
|
|
|
|