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 3394983 queued and waiting for resources srun: job 3394983 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, 0, 0, ..., 106761, 106761, 106762]), col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]), values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379), nnz=106762, layout=torch.sparse_csr) tensor([0.4886, 0.3652, 0.5691, ..., 0.6466, 0.4355, 0.8397]) Matrix: as-caida Shape: torch.Size([31379, 31379]) NNZ: 106762 Density: 0.00010842726485909405 Time: 2.6297245025634766 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/as-caida.mtx 1000': 61.40 msec task-clock:u # 0.010 CPUs utilized 0 context-switches:u # 0.000 /sec 0 cpu-migrations:u # 0.000 /sec 3,507 page-faults:u # 57.117 K/sec 78,967,021 cycles:u # 1.286 GHz (61.13%) 94,334,531 instructions:u # 1.19 insn per cycle (95.16%) branches:u 365,239 branch-misses:u 33,334,312 L1-dcache-loads:u # 542.906 M/sec 457,950 L1-dcache-load-misses:u # 1.37% of all L1-dcache accesses LLC-loads:u LLC-load-misses:u 31,725,851 L1-icache-loads:u # 516.709 M/sec 297,720 L1-icache-load-misses:u # 0.94% of all L1-icache accesses 25,188,580 dTLB-loads:u # 410.239 M/sec (5.16%) dTLB-load-misses:u (0.00%) iTLB-loads:u (0.00%) iTLB-load-misses:u (0.00%) 6.049042045 seconds time elapsed 17.649315000 seconds user 29.335859000 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, 0, 0, ..., 106761, 106761, 106762]), col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]), values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379), nnz=106762, layout=torch.sparse_csr) tensor([0.8344, 0.2588, 0.2246, ..., 0.5607, 0.8141, 0.9893]) Matrix: as-caida Shape: torch.Size([31379, 31379]) NNZ: 106762 Density: 0.00010842726485909405 Time: 2.6495532989501953 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/as-caida.mtx 1000': 325,893 BR_MIS_PRED_RETIRED:u # 0.0 per branch branch_misprediction_ratio 19,069,753 BR_RETIRED:u 6.023780447 seconds time elapsed 17.654658000 seconds user 28.848805000 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, 0, 0, ..., 106761, 106761, 106762]), col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]), values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379), nnz=106762, layout=torch.sparse_csr) tensor([0.0814, 0.1132, 0.8515, ..., 0.8987, 0.5912, 0.5002]) Matrix: as-caida Shape: torch.Size([31379, 31379]) NNZ: 106762 Density: 0.00010842726485909405 Time: 2.5444185733795166 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/as-caida.mtx 1000': 27,181,279 L1I_TLB:u # 0.0 per TLB access itlb_walk_ratio 5,995 ITLB_WALK:u 17,412 DTLB_WALK:u # 0.0 per TLB access dtlb_walk_ratio 37,016,930 L1D_TLB:u 5.790360666 seconds time elapsed 17.919315000 seconds user 30.569858000 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, 0, 0, ..., 106761, 106761, 106762]), col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]), values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379), nnz=106762, layout=torch.sparse_csr) tensor([0.0439, 0.1884, 0.3342, ..., 0.2027, 0.5532, 0.7245]) Matrix: as-caida Shape: torch.Size([31379, 31379]) NNZ: 106762 Density: 0.00010842726485909405 Time: 2.620804786682129 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/as-caida.mtx 1000': 31,535,482 L1I_CACHE:u # 0.0 per cache access l1i_cache_miss_ratio 292,676 L1I_CACHE_REFILL:u 471,752 L1D_CACHE_REFILL:u # 0.0 per cache access l1d_cache_miss_ratio 33,119,145 L1D_CACHE:u 6.002311801 seconds time elapsed 17.427887000 seconds user 30.063688000 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, 0, 0, ..., 106761, 106761, 106762]), col_indices=tensor([ 106, 329, 1040, ..., 155, 160, 12170]), values=tensor([1., 1., 1., ..., 1., 1., 1.]), size=(31379, 31379), nnz=106762, layout=torch.sparse_csr) tensor([0.1495, 0.5856, 0.8600, ..., 0.2101, 0.6229, 0.2019]) Matrix: as-caida Shape: torch.Size([31379, 31379]) NNZ: 106762 Density: 0.00010842726485909405 Time: 2.561279296875 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/as-caida.mtx 1000': 540,894 LL_CACHE_MISS_RD:u # 1.0 per cache access ll_cache_read_miss_ratio 554,700 LL_CACHE_RD:u 191,772 L2D_TLB:u # 0.1 per TLB access l2_tlb_miss_ratio 23,711 L2D_TLB_REFILL:u 306,195 L2D_CACHE_REFILL:u # 0.2 per cache access l2_cache_miss_ratio 1,755,986 L2D_CACHE:u 5.946428572 seconds time elapsed 17.396567000 seconds user 32.141235000 seconds sys