159 lines
9.5 KiB
Plaintext
159 lines
9.5 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 3394992 queued and waiting for resources
|
|
srun: job 3394992 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, 10, 20, ..., 39994, 39994, 39994]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 9711, 10875, 10876]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(10879, 10879),
|
|
nnz=39994, layout=torch.sparse_csr)
|
|
tensor([0.1181, 0.8387, 0.0554, ..., 0.8107, 0.4393, 0.9489])
|
|
Matrix: p2p-Gnutella04
|
|
Shape: torch.Size([10879, 10879])
|
|
NNZ: 39994
|
|
Density: 0.0003379223282393842
|
|
Time: 1.061662197113037 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/p2p-Gnutella04.mtx 1000':
|
|
|
|
50.59 msec task-clock:u # 0.012 CPUs utilized
|
|
0 context-switches:u # 0.000 /sec
|
|
0 cpu-migrations:u # 0.000 /sec
|
|
3,303 page-faults:u # 65.291 K/sec
|
|
51,318,459 cycles:u # 1.014 GHz (59.34%)
|
|
74,705,078 instructions:u # 1.46 insn per cycle (83.02%)
|
|
<not supported> branches:u
|
|
366,825 branch-misses:u
|
|
31,809,194 L1-dcache-loads:u # 628.781 M/sec
|
|
466,198 L1-dcache-load-misses:u # 1.47% of all L1-dcache accesses
|
|
<not supported> LLC-loads:u
|
|
<not supported> LLC-load-misses:u
|
|
30,390,161 L1-icache-loads:u # 600.731 M/sec
|
|
296,270 L1-icache-load-misses:u # 0.97% of all L1-icache accesses
|
|
61,518,375 dTLB-loads:u # 1.216 G/sec (17.94%)
|
|
<not counted> dTLB-load-misses:u (0.00%)
|
|
<not counted> iTLB-loads:u (0.00%)
|
|
<not counted> iTLB-load-misses:u (0.00%)
|
|
|
|
4.302241563 seconds time elapsed
|
|
|
|
16.122298000 seconds user
|
|
29.141140000 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, 10, 20, ..., 39994, 39994, 39994]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 9711, 10875, 10876]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(10879, 10879),
|
|
nnz=39994, layout=torch.sparse_csr)
|
|
tensor([0.7249, 0.8723, 0.3843, ..., 0.2264, 0.4891, 0.9107])
|
|
Matrix: p2p-Gnutella04
|
|
Shape: torch.Size([10879, 10879])
|
|
NNZ: 39994
|
|
Density: 0.0003379223282393842
|
|
Time: 1.0079431533813477 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/p2p-Gnutella04.mtx 1000':
|
|
|
|
328,853 BR_MIS_PRED_RETIRED:u # 0.0 per branch branch_misprediction_ratio
|
|
19,620,312 BR_RETIRED:u
|
|
|
|
4.241400567 seconds time elapsed
|
|
|
|
15.325937000 seconds user
|
|
28.223386000 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, 10, 20, ..., 39994, 39994, 39994]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 9711, 10875, 10876]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(10879, 10879),
|
|
nnz=39994, layout=torch.sparse_csr)
|
|
tensor([0.7608, 0.2449, 0.5322, ..., 0.5547, 0.8659, 0.8437])
|
|
Matrix: p2p-Gnutella04
|
|
Shape: torch.Size([10879, 10879])
|
|
NNZ: 39994
|
|
Density: 0.0003379223282393842
|
|
Time: 1.1017234325408936 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/p2p-Gnutella04.mtx 1000':
|
|
|
|
27,939,682 L1I_TLB:u # 0.0 per TLB access itlb_walk_ratio
|
|
5,470 ITLB_WALK:u
|
|
17,679 DTLB_WALK:u # 0.0 per TLB access dtlb_walk_ratio
|
|
37,425,602 L1D_TLB:u
|
|
|
|
4.296820500 seconds time elapsed
|
|
|
|
15.875162000 seconds user
|
|
28.803412000 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, 10, 20, ..., 39994, 39994, 39994]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 9711, 10875, 10876]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(10879, 10879),
|
|
nnz=39994, layout=torch.sparse_csr)
|
|
tensor([0.9980, 0.9991, 0.6749, ..., 0.4225, 0.7297, 0.3717])
|
|
Matrix: p2p-Gnutella04
|
|
Shape: torch.Size([10879, 10879])
|
|
NNZ: 39994
|
|
Density: 0.0003379223282393842
|
|
Time: 1.0812580585479736 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/p2p-Gnutella04.mtx 1000':
|
|
|
|
30,276,633 L1I_CACHE:u # 0.0 per cache access l1i_cache_miss_ratio
|
|
291,467 L1I_CACHE_REFILL:u
|
|
479,061 L1D_CACHE_REFILL:u # 0.0 per cache access l1d_cache_miss_ratio
|
|
31,689,326 L1D_CACHE:u
|
|
|
|
4.500137840 seconds time elapsed
|
|
|
|
15.794710000 seconds user
|
|
27.773851000 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, 10, 20, ..., 39994, 39994, 39994]),
|
|
col_indices=tensor([ 1, 2, 3, ..., 9711, 10875, 10876]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(10879, 10879),
|
|
nnz=39994, layout=torch.sparse_csr)
|
|
tensor([0.8707, 0.5871, 0.5970, ..., 0.8826, 0.4673, 0.4994])
|
|
Matrix: p2p-Gnutella04
|
|
Shape: torch.Size([10879, 10879])
|
|
NNZ: 39994
|
|
Density: 0.0003379223282393842
|
|
Time: 0.9900743961334229 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/p2p-Gnutella04.mtx 1000':
|
|
|
|
529,426 LL_CACHE_MISS_RD:u # 1.0 per cache access ll_cache_read_miss_ratio
|
|
550,033 LL_CACHE_RD:u
|
|
171,913 L2D_TLB:u # 0.1 per TLB access l2_tlb_miss_ratio
|
|
20,624 L2D_TLB_REFILL:u
|
|
296,662 L2D_CACHE_REFILL:u # 0.2 per cache access l2_cache_miss_ratio
|
|
1,714,211 L2D_CACHE:u
|
|
|
|
4.284402033 seconds time elapsed
|
|
|
|
15.584671000 seconds user
|
|
27.523772000 seconds sys
|
|
|
|
|
|
|