cephi/ampere_research
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
119bc9d8f9
ampere_research/pytorch/output/altra_10_30_de2010_1000.output
cephi 93690abfee new output
2024-12-03 08:53:39 -05:00

169 lines
9.9 KiB
Plaintext
Raw Blame History

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 3394985 queued and waiting for resources
srun: job 3394985 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, 13, 21, ..., 116047, 116051,
116056]),
col_indices=tensor([ 250, 251, 757, ..., 23334, 23553, 24050]),
values=tensor([ 14900., 33341., 20255., ..., 164227., 52413.,
16949.]), size=(24115, 24115), nnz=116056,
layout=torch.sparse_csr)
tensor([0.6055, 0.8789, 0.0482, ..., 0.0736, 0.1316, 0.6744])
Matrix: de2010
Shape: torch.Size([24115, 24115])
NNZ: 116056
Density: 0.0001995689928120616
Time: 2.6956887245178223 seconds
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 1000':
48.96 msec task-clock:u # 0.008 CPUs utilized
0 context-switches:u # 0.000 /sec
0 cpu-migrations:u # 0.000 /sec
3,285 page-faults:u # 67.090 K/sec
48,563,060 cycles:u # 0.992 GHz (59.76%)
73,465,190 instructions:u # 1.51 insn per cycle (78.23%)
<not supported> branches:u
369,314 branch-misses:u (98.16%)
31,769,641 L1-dcache-loads:u # 648.836 M/sec
479,594 L1-dcache-load-misses:u # 1.51% of all L1-dcache accesses
<not supported> LLC-loads:u
<not supported> LLC-load-misses:u
30,338,929 L1-icache-loads:u # 619.616 M/sec
282,162 L1-icache-load-misses:u # 0.93% of all L1-icache accesses
55,516,925 dTLB-loads:u # 1.134 G/sec (23.54%)
12,345 dTLB-load-misses:u # 0.02% of all dTLB cache accesses (3.47%)
<not counted> iTLB-loads:u (0.00%)
<not counted> iTLB-load-misses:u (0.00%)
6.017085179 seconds time elapsed
17.484355000 seconds user
28.678064000 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, 13, 21, ..., 116047, 116051,
116056]),
col_indices=tensor([ 250, 251, 757, ..., 23334, 23553, 24050]),
values=tensor([ 14900., 33341., 20255., ..., 164227., 52413.,
16949.]), size=(24115, 24115), nnz=116056,
layout=torch.sparse_csr)
tensor([0.2815, 0.8196, 0.3706, ..., 0.1328, 0.4062, 0.9113])
Matrix: de2010
Shape: torch.Size([24115, 24115])
NNZ: 116056
Density: 0.0001995689928120616
Time: 2.7908551692962646 seconds
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 1000':
326,361 BR_MIS_PRED_RETIRED:u # 0.0 per branch branch_misprediction_ratio
19,599,354 BR_RETIRED:u
6.215591535 seconds time elapsed
18.097112000 seconds user
27.831633000 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, 13, 21, ..., 116047, 116051,
116056]),
col_indices=tensor([ 250, 251, 757, ..., 23334, 23553, 24050]),
values=tensor([ 14900., 33341., 20255., ..., 164227., 52413.,
16949.]), size=(24115, 24115), nnz=116056,
layout=torch.sparse_csr)
tensor([0.9002, 0.0843, 0.5558, ..., 0.3931, 0.8070, 0.7414])
Matrix: de2010
Shape: torch.Size([24115, 24115])
NNZ: 116056
Density: 0.0001995689928120616
Time: 2.819589376449585 seconds
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 1000':
26,666,488 L1I_TLB:u # 0.0 per TLB access itlb_walk_ratio
6,643 ITLB_WALK:u
17,347 DTLB_WALK:u # 0.0 per TLB access dtlb_walk_ratio
35,986,736 L1D_TLB:u
6.243883495 seconds time elapsed
17.783312000 seconds user
31.714619000 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, 13, 21, ..., 116047, 116051,
116056]),
col_indices=tensor([ 250, 251, 757, ..., 23334, 23553, 24050]),
values=tensor([ 14900., 33341., 20255., ..., 164227., 52413.,
16949.]), size=(24115, 24115), nnz=116056,
layout=torch.sparse_csr)
tensor([0.9109, 0.6392, 0.7899, ..., 0.0945, 0.3298, 0.6865])
Matrix: de2010
Shape: torch.Size([24115, 24115])
NNZ: 116056
Density: 0.0001995689928120616
Time: 2.747800827026367 seconds
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 1000':
32,502,068 L1I_CACHE:u # 0.0 per cache access l1i_cache_miss_ratio
302,739 L1I_CACHE_REFILL:u
480,619 L1D_CACHE_REFILL:u # 0.0 per cache access l1d_cache_miss_ratio
34,031,072 L1D_CACHE:u
6.126767063 seconds time elapsed
17.702029000 seconds user
29.137072000 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, 13, 21, ..., 116047, 116051,
116056]),
col_indices=tensor([ 250, 251, 757, ..., 23334, 23553, 24050]),
values=tensor([ 14900., 33341., 20255., ..., 164227., 52413.,
16949.]), size=(24115, 24115), nnz=116056,
layout=torch.sparse_csr)
tensor([0.7083, 0.6766, 0.7649, ..., 0.3027, 0.9885, 0.8086])
Matrix: de2010
Shape: torch.Size([24115, 24115])
NNZ: 116056
Density: 0.0001995689928120616
Time: 2.795116901397705 seconds
Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 1000':
552,815 LL_CACHE_MISS_RD:u # 1.0 per cache access ll_cache_read_miss_ratio
567,373 LL_CACHE_RD:u
188,248 L2D_TLB:u # 0.1 per TLB access l2_tlb_miss_ratio
23,165 L2D_TLB_REFILL:u
308,211 L2D_CACHE_REFILL:u # 0.2 per cache access l2_cache_miss_ratio
1,787,647 L2D_CACHE:u
6.041792624 seconds time elapsed
17.791735000 seconds user
29.790006000 seconds sys
Reference in New Issue View Git Blame Copy Permalink