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 3394143 queued and waiting for resources srun: job 3394143 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, 4, 7, ..., 155588, 155592, 155598]), col_indices=tensor([ 131, 561, 996, ..., 32237, 32238, 32570]), values=tensor([79040., 7820., 15136., ..., 2828., 17986., 2482.]), size=(32580, 32580), nnz=155598, layout=torch.sparse_csr) tensor([0.9170, 0.7306, 0.1175, ..., 0.0616, 0.0147, 0.6403]) Shape: torch.Size([32580, 32580]) NNZ: 155598 Density: 0.00014658915806621921 Time: 0.4440653324127197 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/vt2010.mtx 100': 61.63 msec task-clock:u # 0.016 CPUs utilized 0 context-switches:u # 0.000 /sec 0 cpu-migrations:u # 0.000 /sec 3,304 page-faults:u # 53.611 K/sec 64,734,203 cycles:u # 1.050 GHz (50.46%) 53,597,991 instructions:u # 0.83 insn per cycle (70.10%) branches:u 347,389 branch-misses:u (91.95%) 31,363,842 L1-dcache-loads:u # 508.915 M/sec 482,780 L1-dcache-load-misses:u # 1.54% of all L1-dcache accesses LLC-loads:u LLC-load-misses:u 30,027,001 L1-icache-loads:u # 487.223 M/sec 288,023 L1-icache-load-misses:u # 0.96% of all L1-icache accesses 44,333,825 dTLB-loads:u # 719.368 M/sec (48.58%) 74,525 dTLB-load-misses:u # 0.17% of all dTLB cache accesses (16.71%) iTLB-loads:u (0.00%) iTLB-load-misses:u (0.00%) 3.811654040 seconds time elapsed 15.616953000 seconds user 30.906234000 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, 4, 7, ..., 155588, 155592, 155598]), col_indices=tensor([ 131, 561, 996, ..., 32237, 32238, 32570]), values=tensor([79040., 7820., 15136., ..., 2828., 17986., 2482.]), size=(32580, 32580), nnz=155598, layout=torch.sparse_csr) tensor([0.5548, 0.3514, 0.6283, ..., 0.5672, 0.1575, 0.4493]) Shape: torch.Size([32580, 32580]) NNZ: 155598 Density: 0.00014658915806621921 Time: 0.44233155250549316 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/vt2010.mtx 100': 330,777 BR_MIS_PRED_RETIRED:u # 0.0 per branch branch_misprediction_ratio 20,357,034 BR_RETIRED:u 3.835342404 seconds time elapsed 15.497637000 seconds user 28.676763000 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, 4, 7, ..., 155588, 155592, 155598]), col_indices=tensor([ 131, 561, 996, ..., 32237, 32238, 32570]), values=tensor([79040., 7820., 15136., ..., 2828., 17986., 2482.]), size=(32580, 32580), nnz=155598, layout=torch.sparse_csr) tensor([0.0953, 0.5790, 0.0112, ..., 0.9540, 0.3173, 0.4731]) Shape: torch.Size([32580, 32580]) NNZ: 155598 Density: 0.00014658915806621921 Time: 0.43302106857299805 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/vt2010.mtx 100': 27,381,387 L1I_TLB:u # 0.0 per TLB access itlb_walk_ratio 6,248 ITLB_WALK:u 17,636 DTLB_WALK:u # 0.0 per TLB access dtlb_walk_ratio 37,436,110 L1D_TLB:u 3.828586094 seconds time elapsed 15.518057000 seconds user 31.389361000 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, 4, 7, ..., 155588, 155592, 155598]), col_indices=tensor([ 131, 561, 996, ..., 32237, 32238, 32570]), values=tensor([79040., 7820., 15136., ..., 2828., 17986., 2482.]), size=(32580, 32580), nnz=155598, layout=torch.sparse_csr) tensor([0.5456, 0.8708, 0.2037, ..., 0.8669, 0.9122, 0.2046]) Shape: torch.Size([32580, 32580]) NNZ: 155598 Density: 0.00014658915806621921 Time: 0.4426534175872803 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/vt2010.mtx 100': 32,505,993 L1I_CACHE:u # 0.0 per cache access l1i_cache_miss_ratio 303,849 L1I_CACHE_REFILL:u 467,426 L1D_CACHE_REFILL:u # 0.0 per cache access l1d_cache_miss_ratio 34,241,110 L1D_CACHE:u 3.811299200 seconds time elapsed 15.932195000 seconds user 30.887870000 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, 4, 7, ..., 155588, 155592, 155598]), col_indices=tensor([ 131, 561, 996, ..., 32237, 32238, 32570]), values=tensor([79040., 7820., 15136., ..., 2828., 17986., 2482.]), size=(32580, 32580), nnz=155598, layout=torch.sparse_csr) tensor([0.5024, 0.2304, 0.7925, ..., 0.1397, 0.5558, 0.6450]) Shape: torch.Size([32580, 32580]) NNZ: 155598 Density: 0.00014658915806621921 Time: 0.3671383857727051 seconds Performance counter stats for 'apptainer run pytorch-altra.sif -c numactl --cpunodebind=0 --membind=0 python spmv.py matrices/vt2010.mtx 100': 550,075 LL_CACHE_MISS_RD:u # 1.0 per cache access ll_cache_read_miss_ratio 562,829 LL_CACHE_RD:u 199,285 L2D_TLB:u # 0.1 per TLB access l2_tlb_miss_ratio 24,424 L2D_TLB_REFILL:u 310,155 L2D_CACHE_REFILL:u # 0.2 per cache access l2_cache_miss_ratio 1,783,824 L2D_CACHE:u 3.824434783 seconds time elapsed 15.754438000 seconds user 28.226523000 seconds sys