159 lines
9.4 KiB
Plaintext
159 lines
9.4 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 3394980 queued and waiting for resources
|
|
srun: job 3394980 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, 583, 584, ..., 65459, 65460, 65460]),
|
|
col_indices=tensor([ 2, 23, 27, ..., 3324, 958, 841]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(11806, 11806),
|
|
nnz=65460, layout=torch.sparse_csr)
|
|
tensor([0.9231, 0.7723, 0.0509, ..., 0.0839, 0.6982, 0.3459])
|
|
Matrix: Oregon-2
|
|
Shape: torch.Size([11806, 11806])
|
|
NNZ: 65460
|
|
Density: 0.0004696458003979807
|
|
Time: 1.5677142143249512 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/Oregon-2.mtx 1000':
|
|
|
|
64.81 msec task-clock:u # 0.013 CPUs utilized
|
|
0 context-switches:u # 0.000 /sec
|
|
0 cpu-migrations:u # 0.000 /sec
|
|
3,244 page-faults:u # 50.056 K/sec
|
|
82,069,432 cycles:u # 1.266 GHz (59.04%)
|
|
78,292,700 instructions:u # 0.95 insn per cycle (76.75%)
|
|
<not supported> branches:u
|
|
341,509 branch-misses:u (90.97%)
|
|
33,032,555 L1-dcache-loads:u # 509.704 M/sec
|
|
478,674 L1-dcache-load-misses:u # 1.45% of all L1-dcache accesses
|
|
<not supported> LLC-loads:u
|
|
<not supported> LLC-load-misses:u
|
|
31,508,310 L1-icache-loads:u # 486.184 M/sec
|
|
297,528 L1-icache-load-misses:u # 0.94% of all L1-icache accesses
|
|
49,358,091 dTLB-loads:u # 761.613 M/sec (27.83%)
|
|
88,514 dTLB-load-misses:u # 0.18% of all dTLB cache accesses (14.82%)
|
|
<not counted> iTLB-loads:u (0.00%)
|
|
<not counted> iTLB-load-misses:u (0.00%)
|
|
|
|
5.016393105 seconds time elapsed
|
|
|
|
16.759527000 seconds user
|
|
31.429551000 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, 583, 584, ..., 65459, 65460, 65460]),
|
|
col_indices=tensor([ 2, 23, 27, ..., 3324, 958, 841]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(11806, 11806),
|
|
nnz=65460, layout=torch.sparse_csr)
|
|
tensor([0.8423, 0.9339, 0.8037, ..., 0.5953, 0.0649, 0.1559])
|
|
Matrix: Oregon-2
|
|
Shape: torch.Size([11806, 11806])
|
|
NNZ: 65460
|
|
Density: 0.0004696458003979807
|
|
Time: 1.516484022140503 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/Oregon-2.mtx 1000':
|
|
|
|
319,703 BR_MIS_PRED_RETIRED:u # 0.0 per branch branch_misprediction_ratio
|
|
19,996,903 BR_RETIRED:u
|
|
|
|
4.945699041 seconds time elapsed
|
|
|
|
16.431978000 seconds user
|
|
29.752452000 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, 583, 584, ..., 65459, 65460, 65460]),
|
|
col_indices=tensor([ 2, 23, 27, ..., 3324, 958, 841]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(11806, 11806),
|
|
nnz=65460, layout=torch.sparse_csr)
|
|
tensor([0.8058, 0.2922, 0.1227, ..., 0.2176, 0.9496, 0.8838])
|
|
Matrix: Oregon-2
|
|
Shape: torch.Size([11806, 11806])
|
|
NNZ: 65460
|
|
Density: 0.0004696458003979807
|
|
Time: 1.6458909511566162 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/Oregon-2.mtx 1000':
|
|
|
|
26,988,315 L1I_TLB:u # 0.0 per TLB access itlb_walk_ratio
|
|
5,988 ITLB_WALK:u
|
|
14,570 DTLB_WALK:u # 0.0 per TLB access dtlb_walk_ratio
|
|
36,879,854 L1D_TLB:u
|
|
|
|
5.011871473 seconds time elapsed
|
|
|
|
16.529942000 seconds user
|
|
30.438432000 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, 583, 584, ..., 65459, 65460, 65460]),
|
|
col_indices=tensor([ 2, 23, 27, ..., 3324, 958, 841]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(11806, 11806),
|
|
nnz=65460, layout=torch.sparse_csr)
|
|
tensor([0.7728, 0.1182, 0.3337, ..., 0.2555, 0.2523, 0.5746])
|
|
Matrix: Oregon-2
|
|
Shape: torch.Size([11806, 11806])
|
|
NNZ: 65460
|
|
Density: 0.0004696458003979807
|
|
Time: 1.529954433441162 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/Oregon-2.mtx 1000':
|
|
|
|
30,465,174 L1I_CACHE:u # 0.0 per cache access l1i_cache_miss_ratio
|
|
293,085 L1I_CACHE_REFILL:u
|
|
487,330 L1D_CACHE_REFILL:u # 0.0 per cache access l1d_cache_miss_ratio
|
|
31,932,249 L1D_CACHE:u
|
|
|
|
4.954100105 seconds time elapsed
|
|
|
|
16.282966000 seconds user
|
|
28.926724000 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, 583, 584, ..., 65459, 65460, 65460]),
|
|
col_indices=tensor([ 2, 23, 27, ..., 3324, 958, 841]),
|
|
values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(11806, 11806),
|
|
nnz=65460, layout=torch.sparse_csr)
|
|
tensor([0.5613, 0.3211, 0.1739, ..., 0.5461, 0.1391, 0.8387])
|
|
Matrix: Oregon-2
|
|
Shape: torch.Size([11806, 11806])
|
|
NNZ: 65460
|
|
Density: 0.0004696458003979807
|
|
Time: 1.5726752281188965 seconds
|
|
|
|
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/Oregon-2.mtx 1000':
|
|
|
|
545,501 LL_CACHE_MISS_RD:u # 1.0 per cache access ll_cache_read_miss_ratio
|
|
558,084 LL_CACHE_RD:u
|
|
204,746 L2D_TLB:u # 0.1 per TLB access l2_tlb_miss_ratio
|
|
25,302 L2D_TLB_REFILL:u
|
|
314,594 L2D_CACHE_REFILL:u # 0.2 per cache access l2_cache_miss_ratio
|
|
1,828,047 L2D_CACHE:u
|
|
|
|
4.866549675 seconds time elapsed
|
|
|
|
16.609257000 seconds user
|
|
31.381282000 seconds sys
|
|
|
|
|
|
|