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 3394139 queued and waiting for resources srun: job 3394139 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([0.4207, 0.3943, 0.6543, ..., 0.2191, 0.5415, 0.1575]) Shape: torch.Size([24115, 24115]) NNZ: 116056 Density: 0.0001995689928120616 Time: 0.36042284965515137 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 100': 59.88 msec task-clock:u # 0.016 CPUs utilized 0 context-switches:u # 0.000 /sec 0 cpu-migrations:u # 0.000 /sec 3,313 page-faults:u # 55.328 K/sec 58,169,777 cycles:u # 0.971 GHz (61.49%) 57,993,431 instructions:u # 1.00 insn per cycle (81.67%) branches:u 341,266 branch-misses:u 31,858,781 L1-dcache-loads:u # 532.049 M/sec 467,486 L1-dcache-load-misses:u # 1.47% of all L1-dcache accesses LLC-loads:u LLC-load-misses:u 30,461,310 L1-icache-loads:u # 508.711 M/sec 294,156 L1-icache-load-misses:u # 0.97% of all L1-icache accesses 43,828,130 dTLB-loads:u # 731.940 M/sec (40.26%) 47,836 dTLB-load-misses:u # 0.11% of all dTLB cache accesses (25.52%) 0 iTLB-loads:u # 0.000 /sec (2.73%) iTLB-load-misses:u (0.00%) 3.824054028 seconds time elapsed 15.099361000 seconds user 28.830417000 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([0.0456, 0.2095, 0.0276, ..., 0.4209, 0.6824, 0.5475]) Shape: torch.Size([24115, 24115]) NNZ: 116056 Density: 0.0001995689928120616 Time: 0.3598823547363281 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 100': 330,494 BR_MIS_PRED_RETIRED:u # 0.0 per branch branch_misprediction_ratio 20,578,427 BR_RETIRED:u 3.781234836 seconds time elapsed 14.965545000 seconds user 29.444131000 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([0.9882, 0.5477, 0.6307, ..., 0.1179, 0.6903, 0.1235]) Shape: torch.Size([24115, 24115]) NNZ: 116056 Density: 0.0001995689928120616 Time: 0.29088521003723145 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 100': 27,982,097 L1I_TLB:u # 0.0 per TLB access itlb_walk_ratio 6,614 ITLB_WALK:u 17,270 DTLB_WALK:u # 0.0 per TLB access dtlb_walk_ratio 37,728,899 L1D_TLB:u 3.576632300 seconds time elapsed 14.864601000 seconds user 29.274547000 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([0.3952, 0.0475, 0.1125, ..., 0.3481, 0.1290, 0.3495]) Shape: torch.Size([24115, 24115]) NNZ: 116056 Density: 0.0001995689928120616 Time: 0.30365920066833496 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 100': 29,754,926 L1I_CACHE:u # 0.0 per cache access l1i_cache_miss_ratio 278,786 L1I_CACHE_REFILL:u 454,742 L1D_CACHE_REFILL:u # 0.0 per cache access l1d_cache_miss_ratio 31,173,246 L1D_CACHE:u 3.730995381 seconds time elapsed 15.213930000 seconds user 30.995070000 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([0.7266, 0.7537, 0.9729, ..., 0.3349, 0.3523, 0.6532]) Shape: torch.Size([24115, 24115]) NNZ: 116056 Density: 0.0001995689928120616 Time: 0.2798902988433838 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/de2010.mtx 100': 543,243 LL_CACHE_MISS_RD:u # 1.0 per cache access ll_cache_read_miss_ratio 560,716 LL_CACHE_RD:u 162,281 L2D_TLB:u # 0.1 per TLB access l2_tlb_miss_ratio 19,847 L2D_TLB_REFILL:u 300,577 L2D_CACHE_REFILL:u # 0.2 per cache access l2_cache_miss_ratio 1,696,278 L2D_CACHE:u 3.819959836 seconds time elapsed 15.346035000 seconds user 29.199873000 seconds sys