StartDate: 2021-04-20 22:02:06+00:00 CpuId: 12x Intel Xeon W 2000 / Scalable Bronze 3000 / Silver 4000 / Gold 5000 / 6000 / Platinum 8000 (Skylake), 14nm GpuId: 1x Tesla V100-SXM2-16GB CommitSHA: 4ceb55d7728e483bd4825e6faf97f65e69b001d5 CommitTime: 2021-04-20 23:57:53 +0200 CommitAuthor: Ole Schütt CommitSubject: grid: Refactor GPU backend Trying to pull image cp2k-toolchain-cuda-volta... success :-) Trying to pull image cp2k-perf-cuda-volta... success :-) #################### Running Image cp2k-perf-cuda-volta #################### ========== Fetching Git Commit ========== CommitSHA: 4ceb55d7728e483bd4825e6faf97f65e69b001d5 CommitTime: 2021-04-20 23:57:53 +0200 CommitAuthor: Ole Schütt CommitSubject: grid: Refactor GPU backend ========== Running Test ========== ========== Compiling CP2K ========== Compiling cp2k... done. ========== Running Performance Test ========== Running H2O-64.inp with 3 threads and 2 ranks... done. From /workspace/artifacts/H2O-64_6cpu_1gpu.out: ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.029 0.030 103.147 103.147 qs_mol_dyn_low 1 2.0 0.004 0.004 102.702 102.705 qs_forces 11 3.9 0.001 0.001 102.649 102.649 qs_energies 11 4.9 0.001 0.001 89.363 89.365 scf_env_do_scf 11 5.9 0.001 0.001 70.813 70.814 velocity_verlet 10 3.0 0.001 0.002 66.056 66.075 scf_env_do_scf_inner_loop 108 6.5 0.009 0.013 55.483 55.483 dbcsr_multiply_generic 2286 12.5 0.138 0.138 27.001 27.065 rebuild_ks_matrix 119 8.3 0.001 0.001 25.118 25.121 qs_ks_build_kohn_sham_matrix 119 9.3 0.018 0.018 25.117 25.120 qs_ks_update_qs_env 119 7.6 0.001 0.001 22.334 22.336 qs_scf_new_mos 108 7.5 0.001 0.001 21.173 21.192 qs_scf_loop_do_ot 108 8.5 0.001 0.001 21.172 21.191 ot_scf_mini 108 9.5 0.003 0.003 16.234 16.235 qs_rho_update_rho 119 7.7 0.001 0.001 15.731 15.761 calculate_rho_elec 119 8.7 0.969 0.975 15.731 15.760 init_scf_loop 11 6.9 0.000 0.000 15.254 15.254 pw_transfer 1439 11.6 0.094 0.095 14.368 14.457 fft_wrap_pw1pw2 1201 12.6 0.010 0.011 13.952 14.042 sum_up_and_integrate 119 10.3 0.178 0.179 13.438 13.510 integrate_v_rspace 119 11.3 0.367 0.368 13.259 13.332 multiply_cannon 2286 13.5 0.448 0.450 13.169 13.205 prepare_preconditioner 11 7.9 0.000 0.000 12.476 12.479 make_preconditioner 11 8.9 0.000 0.000 12.476 12.479 fft_wrap_pw1pw2_140 487 13.2 0.002 0.002 11.841 11.982 make_full_inverse_cholesky 11 9.9 0.000 0.000 11.813 11.821 multiply_cannon_loop 2286 14.5 0.167 0.167 11.802 11.817 qs_energies_init_hamiltonians 11 5.9 0.000 0.000 10.077 10.078 multiply_cannon_multrec 4572 15.5 1.725 1.750 9.839 9.847 ot_mini 108 10.5 0.001 0.001 9.706 9.709 build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 9.458 9.612 density_rs2pw 119 9.7 0.007 0.007 9.467 9.594 make_m2s 4572 13.5 0.047 0.048 8.245 8.275 make_images 4572 14.5 1.459 1.462 8.068 8.096 cp_fm_cholesky_invert 11 10.9 7.976 7.976 7.976 7.976 init_scf_run 11 5.9 0.001 0.001 7.482 7.482 scf_env_initial_rho_setup 11 6.9 0.000 0.001 7.481 7.482 grid_integrate_task_list 119 12.3 7.316 7.387 7.316 7.387 calculate_dm_sparse 119 9.5 0.001 0.001 7.261 7.279 pw_cuda_r3dc1d_3d_ps 606 14.1 0.010 0.011 7.256 7.279 pw_cuda_c1dr3d_3d_ps 595 15.2 0.010 0.011 6.681 6.748 qs_ot_get_derivative 108 11.5 0.001 0.001 6.247 6.247 potential_pw2rs 119 12.3 0.038 0.038 5.577 5.579 dbcsr_mm_accdrv_process 9594 16.2 1.173 1.261 5.368 5.368 grid_collocate_task_list 119 9.7 5.228 5.283 5.228 5.283 qs_env_update_s_mstruct 11 6.9 0.000 0.000 5.104 5.129 build_core_ppl_forces 11 5.9 4.757 4.878 4.757 4.878 build_core_hamiltonian_matrix 11 6.9 0.001 0.001 4.609 4.664 rs_pw_transfer 974 11.9 0.017 0.018 4.374 4.518 dbcsr_make_untransposed_blocks 2729 13.4 4.178 4.191 4.191 4.204 qs_create_task_list 11 7.9 0.000 0.000 4.081 4.118 generate_qs_task_list 11 8.9 1.889 1.908 4.080 4.118 wfi_extrapolate 11 7.9 0.001 0.001 3.724 3.724 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.654 3.654 calculate_first_density_matrix 1 7.0 0.000 0.000 3.570 3.571 jit_kernel_multiply 11 15.6 3.414 3.498 3.414 3.498 ot_diis_step 108 11.5 0.006 0.006 3.443 3.443 hybrid_alltoall_any 4725 16.4 2.021 2.034 3.154 3.178 make_images_data 4572 15.5 0.050 0.051 3.092 3.105 cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.087 3.087 pw_cuda_fg 606 15.1 2.994 3.044 2.994 3.044 cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 2.945 2.946 mp_alltoall_z22v 1201 16.6 2.840 2.915 2.840 2.915 apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.893 2.898 apply_single 119 13.6 0.000 0.001 2.893 2.898 mp_waitall_1 64982 16.9 2.735 2.867 2.735 2.867 qs_ot_get_p 119 10.4 0.001 0.001 2.790 2.794 multiply_cannon_multrec_finali 2286 16.5 0.004 0.004 2.747 2.765 dbcsr_mm_multrec_finalize 2286 17.5 0.024 0.025 2.743 2.761 dbcsr_mm_sched_finalize 2286 18.5 2.716 2.733 2.719 2.736 pw_cuda_sf 595 16.2 2.712 2.719 2.712 2.719 build_core_ppl 11 7.9 2.646 2.699 2.646 2.699 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.596 2.598 dbcsr_complete_redistribute 329 12.2 1.289 1.324 2.509 2.538 qs_vxc_create 119 10.3 0.002 0.002 2.378 2.428 xc_vxc_pw_create 119 11.3 0.692 0.695 2.376 2.426 rs_pw_transfer_RS2PW_140 130 11.5 1.658 1.672 2.248 2.403 pw_poisson_solve 119 10.3 0.826 0.827 2.385 2.385 build_kinetic_matrix_low 22 6.9 2.233 2.244 2.335 2.344 build_overlap_matrix_low 22 6.9 2.197 2.200 2.297 2.302 qs_ot_get_derivative_taylor 59 13.0 0.002 0.002 2.277 2.290 pw_cuda_cff 606 15.1 2.279 2.284 2.279 2.284 dbcsr_special_finalize 6858 15.5 0.033 0.033 2.199 2.204 dbcsr_merge_single_wm 4572 16.5 0.146 0.150 2.098 2.104 copy_dbcsr_to_fm 153 11.3 0.003 0.003 2.069 2.093 pw_cuda_ffc 595 16.2 1.992 2.082 1.992 2.082 ------------------------------------------------------------------------------- Plot: name="H2O-64_timings_6cpu_1gpu", title="Timings of H2O-64 with 6 CPU Cores and 1 GPU", ylabel="time [s]" PlotPoint: plot="H2O-64_timings_6cpu_1gpu", name="rest", label="rest", y=73.69200000000001, yerr=0.0 PlotPoint: plot="H2O-64_timings_6cpu_1gpu", name="cp_fm_cholesky_invert", label="cp_fm_cholesky_invert", y=7.976, yerr=0.0 PlotPoint: plot="H2O-64_timings_6cpu_1gpu", name="grid_integrate_task_list", label="grid_integrate_task_list", y=7.316, yerr=0.0 PlotPoint: plot="H2O-64_timings_6cpu_1gpu", name="grid_collocate_task_list", label="grid_collocate_task_list", y=5.228, yerr=0.0 PlotPoint: plot="H2O-64_timings_6cpu_1gpu", name="build_core_ppl_forces", label="build_core_ppl_forces", y=4.757, yerr=0.0 PlotPoint: plot="H2O-64_timings_6cpu_1gpu", name="dbcsr_make_untransposed_blocks", label="dbcsr_make_untransposed_blocks", y=4.178, yerr=0.0 Running H2O-64_nonortho.inp with 3 threads and 2 ranks... done. From /workspace/artifacts/H2O-64_nonortho_6cpu_1gpu.out: ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.027 0.028 100.742 100.743 qs_mol_dyn_low 1 2.0 0.004 0.004 100.266 100.269 qs_forces 11 3.9 0.001 0.001 100.212 100.212 qs_energies 11 4.9 0.001 0.001 86.389 86.390 scf_env_do_scf 11 5.9 0.001 0.001 67.973 67.974 velocity_verlet 10 3.0 0.001 0.002 66.596 66.616 scf_env_do_scf_inner_loop 96 6.5 0.008 0.011 52.485 52.486 rebuild_ks_matrix 107 8.3 0.001 0.001 24.736 24.737 qs_ks_build_kohn_sham_matrix 107 9.3 0.016 0.016 24.736 24.736 dbcsr_multiply_generic 1966 12.4 0.120 0.121 23.277 23.314 qs_ks_update_qs_env 107 7.6 0.001 0.001 21.444 21.446 qs_scf_new_mos 96 7.5 0.001 0.001 18.241 18.242 qs_scf_loop_do_ot 96 8.5 0.001 0.001 18.240 18.241 qs_rho_update_rho 107 7.7 0.001 0.001 17.036 17.058 calculate_rho_elec 107 8.7 0.857 0.867 17.035 17.057 init_scf_loop 11 6.9 0.000 0.000 15.411 15.411 sum_up_and_integrate 107 10.3 0.159 0.162 14.246 14.275 integrate_v_rspace 107 11.3 0.331 0.332 14.086 14.118 ot_scf_mini 96 9.5 0.003 0.003 13.930 13.930 pw_transfer 1295 11.6 0.085 0.085 12.964 12.968 fft_wrap_pw1pw2 1081 12.6 0.009 0.009 12.590 12.593 prepare_preconditioner 11 7.9 0.000 0.000 12.457 12.458 make_preconditioner 11 8.9 0.000 0.000 12.457 12.458 make_full_inverse_cholesky 11 9.9 0.000 0.000 11.767 11.774 multiply_cannon 1966 13.4 0.396 0.400 10.876 10.889 fft_wrap_pw1pw2_140 439 13.2 0.002 0.002 10.680 10.724 qs_energies_init_hamiltonians 11 5.9 0.000 0.000 10.591 10.591 build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 9.561 9.746 multiply_cannon_loop 1966 14.4 0.143 0.146 9.685 9.691 grid_integrate_task_list 107 12.3 8.712 8.742 8.712 8.742 density_rs2pw 107 9.7 0.006 0.006 8.496 8.570 ot_mini 96 10.5 0.001 0.001 8.241 8.241 multiply_cannon_multrec 3932 15.4 1.525 1.553 8.147 8.170 cp_fm_cholesky_invert 11 10.9 8.036 8.036 8.036 8.036 grid_collocate_task_list 107 9.7 7.623 7.677 7.623 7.677 make_m2s 3932 13.4 0.041 0.041 7.405 7.408 make_images 3932 14.4 1.306 1.311 7.249 7.252 init_scf_run 11 5.9 0.001 0.001 6.833 6.833 scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.833 6.833 pw_cuda_r3dc1d_3d_ps 546 14.1 0.009 0.009 6.576 6.598 calculate_dm_sparse 107 9.5 0.001 0.001 6.104 6.105 pw_cuda_c1dr3d_3d_ps 535 15.2 0.009 0.010 6.000 6.020 qs_env_update_s_mstruct 11 6.9 0.000 0.000 5.644 5.683 qs_ot_get_derivative 96 11.5 0.001 0.001 5.188 5.189 potential_pw2rs 107 12.3 0.034 0.034 5.043 5.046 build_core_ppl_forces 11 5.9 4.786 4.918 4.786 4.918 build_core_hamiltonian_matrix 11 6.9 0.001 0.001 4.596 4.664 qs_create_task_list 11 7.9 0.000 0.000 4.627 4.652 generate_qs_task_list 11 8.9 2.464 2.483 4.627 4.652 rs_pw_transfer 878 11.9 0.015 0.016 3.969 4.110 dbcsr_mm_accdrv_process 8450 16.1 0.748 0.753 4.049 4.074 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 4.072 4.072 wfi_extrapolate 11 7.9 0.001 0.001 3.983 3.983 dbcsr_make_untransposed_blocks 2361 13.3 3.741 3.747 3.752 3.758 ot_diis_step 96 11.5 0.005 0.005 3.039 3.039 hybrid_alltoall_any 4079 16.3 1.837 1.841 2.860 2.864 make_images_data 3932 15.4 0.044 0.044 2.792 2.796 pw_cuda_fg 546 15.1 2.709 2.755 2.709 2.755 build_core_ppl 11 7.9 2.653 2.713 2.653 2.713 cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.656 2.661 calculate_first_density_matrix 1 7.0 0.000 0.000 2.630 2.630 apply_preconditioner_dbcsr 107 12.6 0.000 0.000 2.624 2.624 apply_single 107 13.6 0.000 0.000 2.624 2.624 jit_kernel_multiply 12 15.8 2.602 2.619 2.602 2.619 multiply_cannon_multrec_finali 1966 16.4 0.004 0.004 2.574 2.598 dbcsr_mm_multrec_finalize 1966 17.4 0.022 0.022 2.570 2.595 dbcsr_mm_sched_finalize 1966 18.4 2.546 2.570 2.549 2.573 mp_waitall_1 55926 16.8 2.441 2.538 2.441 2.538 dbcsr_complete_redistribute 317 12.2 1.261 1.307 2.505 2.527 mp_alltoall_z22v 1081 16.6 2.523 2.525 2.523 2.525 pw_cuda_sf 535 16.2 2.451 2.460 2.451 2.460 cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 2.416 2.417 build_kinetic_matrix_low 22 6.9 2.265 2.279 2.366 2.380 qs_ot_get_p 107 10.4 0.001 0.001 2.355 2.357 build_overlap_matrix_low 22 6.9 2.207 2.224 2.311 2.331 rs_pw_transfer_RS2PW_140 118 11.5 1.498 1.521 2.029 2.183 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.161 2.162 pw_poisson_solve 107 10.3 0.747 0.747 2.141 2.141 pw_cuda_cff 546 15.1 2.083 2.116 2.083 2.116 qs_vxc_create 107 10.3 0.002 0.002 2.095 2.096 xc_vxc_pw_create 107 11.3 0.617 0.623 2.093 2.094 copy_dbcsr_to_fm 147 11.2 0.003 0.003 2.010 2.021 ------------------------------------------------------------------------------- Plot: name="H2O-64_nonortho_timings_6cpu_1gpu", title="Timings of H2O-64_nonortho with 6 CPU Cores and 1 GPU", ylabel="time [s]" PlotPoint: plot="H2O-64_nonortho_timings_6cpu_1gpu", name="rest", label="rest", y=67.84400000000001, yerr=0.0 PlotPoint: plot="H2O-64_nonortho_timings_6cpu_1gpu", name="grid_integrate_task_list", label="grid_integrate_task_list", y=8.712, yerr=0.0 PlotPoint: plot="H2O-64_nonortho_timings_6cpu_1gpu", name="cp_fm_cholesky_invert", label="cp_fm_cholesky_invert", y=8.036, yerr=0.0 PlotPoint: plot="H2O-64_nonortho_timings_6cpu_1gpu", name="grid_collocate_task_list", label="grid_collocate_task_list", y=7.623, yerr=0.0 PlotPoint: plot="H2O-64_nonortho_timings_6cpu_1gpu", name="build_core_ppl_forces", label="build_core_ppl_forces", y=4.786, yerr=0.0 PlotPoint: plot="H2O-64_nonortho_timings_6cpu_1gpu", name="dbcsr_make_untransposed_blocks", label="dbcsr_make_untransposed_blocks", y=3.741, yerr=0.0 Running bench_dftb.inp with 3 threads and 2 ranks... done. From /workspace/artifacts/bench_dftb_6cpu_1gpu.out: ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.053 0.054 268.200 268.200 qs_energies 1 2.0 0.000 0.000 268.081 268.082 ls_scf 1 3.0 0.000 0.000 267.122 267.129 ls_scf_main 1 4.0 0.001 0.002 257.137 257.138 density_matrix_trs4 11 5.0 0.009 0.011 216.507 216.509 dbcsr_multiply_generic 185 6.1 0.302 0.303 178.524 178.560 multiply_cannon 185 7.1 0.993 1.094 127.109 127.142 multiply_cannon_loop 185 8.1 0.300 0.301 111.436 111.636 multiply_cannon_multrec 370 9.1 94.098 94.424 106.301 106.513 make_m2s 370 7.1 0.028 0.028 41.086 41.130 make_images 370 8.1 13.829 14.258 40.093 40.132 ls_scf_dm_to_ks 11 5.0 0.000 0.000 36.629 36.645 matrix_ls_to_qs 11 6.0 0.000 0.000 34.338 34.369 dbcsr_complete_redistribute 23 7.5 21.423 21.536 29.860 29.906 matrix_decluster 11 7.0 0.000 0.000 27.443 27.487 arnoldi_extremal 12 6.1 0.000 0.000 26.612 26.613 arnoldi_normal_ev 12 7.1 0.013 0.014 26.612 26.612 build_subspace 23 8.1 0.064 0.064 26.059 26.059 dbcsr_matrix_vector_mult 652 9.0 0.129 0.129 25.026 25.048 dbcsr_matrix_vector_mult_local 652 10.0 24.036 24.058 24.041 24.063 dbcsr_finalize 646 7.5 0.204 0.211 15.287 15.291 dbcsr_merge_all 597 8.5 1.914 1.949 12.622 12.635 setup_rec_index_2d 370 8.1 12.263 12.276 12.263 12.276 make_images_data 370 9.1 0.011 0.011 9.493 9.799 hybrid_alltoall_any 393 9.9 4.582 4.643 9.090 9.392 dbcsr_special_finalize 555 9.1 0.008 0.008 9.361 9.361 dbcsr_sort_indices 1287 10.0 9.043 9.067 9.043 9.067 ls_scf_init_scf 1 4.0 0.000 0.000 8.586 8.592 dbcsr_mm_accdrv_process 14501 10.0 0.583 0.593 8.468 8.518 dbcsr_add_d 280 6.0 0.001 0.001 8.224 8.270 dbcsr_add_anytype 280 7.0 4.192 4.195 8.223 8.269 ls_scf_init_matrix_S 1 5.0 0.000 0.000 8.061 8.066 dbcsr_mm_accdrv_process_sort 14501 11.0 7.628 7.669 7.628 7.669 matrix_sqrt_Newton_Schulz 1 6.0 0.001 0.001 7.353 7.355 tree_to_linear_d 110 9.4 7.133 7.144 7.133 7.144 dbcsr_copy_into_existing 11 7.0 6.894 6.969 6.894 6.970 make_images_pack 370 9.1 6.256 6.326 6.275 6.345 dbcsr_merge_single_wm 370 10.1 0.546 0.565 6.000 6.013 ------------------------------------------------------------------------------- Plot: name="bench_dftb_timings_6cpu_1gpu", title="Timings of bench_dftb with 6 CPU Cores and 1 GPU", ylabel="time [s]" PlotPoint: plot="bench_dftb_timings_6cpu_1gpu", name="rest", label="rest", y=102.55099999999999, yerr=0.0 PlotPoint: plot="bench_dftb_timings_6cpu_1gpu", name="multiply_cannon_multrec", label="multiply_cannon_multrec", y=94.098, yerr=0.0 PlotPoint: plot="bench_dftb_timings_6cpu_1gpu", name="dbcsr_matrix_vector_mult_local", label="dbcsr_matrix_vector_mult_local", y=24.036, yerr=0.0 PlotPoint: plot="bench_dftb_timings_6cpu_1gpu", name="dbcsr_complete_redistribute", label="dbcsr_complete_redistribute", y=21.423, yerr=0.0 PlotPoint: plot="bench_dftb_timings_6cpu_1gpu", name="make_images", label="make_images", y=13.829, yerr=0.0 PlotPoint: plot="bench_dftb_timings_6cpu_1gpu", name="setup_rec_index_2d", label="setup_rec_index_2d", y=12.263, yerr=0.0 Running dbcsr.inp with 3 threads and 2 ranks... done. From /workspace/artifacts/dbcsr_6cpu_1gpu.out: ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.008 0.009 47.767 47.768 lib_test 1 2.0 0.000 0.000 47.756 47.758 dbcsr_run_tests 3 3.0 0.001 0.001 47.756 47.757 test_multiplies_multiproc 3 4.0 0.001 0.001 40.346 40.428 dbcsr_multiply_generic 9 5.0 0.001 0.001 29.429 29.433 multiply_cannon 9 6.0 0.204 0.268 21.967 22.235 multiply_cannon_loop 9 7.0 0.003 0.004 20.818 21.045 multiply_cannon_multrec 18 8.0 10.330 10.547 19.525 19.782 dbcsr_mm_accdrv_process 8188 9.0 1.132 1.139 8.364 8.401 dbcsr_redistribute 9 5.0 6.170 6.180 7.960 7.967 dbcsr_make_random_matrix 9 4.0 4.465 4.567 7.338 7.418 dbcsr_finalize 27 5.7 0.001 0.001 6.284 6.466 dbcsr_merge_all 18 6.5 2.730 2.736 6.044 6.228 dbcsr_mm_accdrv_process_sort 8188 10.0 5.328 5.363 5.328 5.363 make_m2s 18 6.0 0.001 0.001 3.461 3.463 make_images 18 7.0 0.475 0.478 3.429 3.431 jit_kernel_multiply 7 10.0 1.903 1.907 1.903 1.907 dbcsr_checksum 6 5.0 1.895 1.896 1.900 1.900 dbcsr_data_copy_aa2 18 7.5 1.554 1.740 1.554 1.740 tree_to_linear_d 9 7.0 1.630 1.636 1.630 1.636 mp_alltoall_d11v 27 6.0 1.621 1.629 1.621 1.629 make_images_data 18 8.0 0.000 0.000 1.133 1.142 acc_transpose_blocks 18 8.0 0.000 0.000 1.111 1.141 acc_transpose_blocks_kernels 18 9.0 0.004 0.004 1.076 1.109 hybrid_alltoall_any 18 9.0 0.754 0.756 1.100 1.109 jit_kernel_transpose 6 10.0 1.072 1.105 1.072 1.105 make_images_pack 18 8.0 1.075 1.078 1.075 1.079 dbcsr_data_release 507 7.7 0.998 1.002 0.998 1.002 ------------------------------------------------------------------------------- Plot: name="dbcsr_timings_6cpu_1gpu", title="Timings of dbcsr with 6 CPU Cores and 1 GPU", ylabel="time [s]" PlotPoint: plot="dbcsr_timings_6cpu_1gpu", name="rest", label="rest", y=18.744000000000003, yerr=0.0 PlotPoint: plot="dbcsr_timings_6cpu_1gpu", name="multiply_cannon_multrec", label="multiply_cannon_multrec", y=10.33, yerr=0.0 PlotPoint: plot="dbcsr_timings_6cpu_1gpu", name="dbcsr_redistribute", label="dbcsr_redistribute", y=6.17, yerr=0.0 PlotPoint: plot="dbcsr_timings_6cpu_1gpu", name="dbcsr_mm_accdrv_process_sort", label="dbcsr_mm_accdrv_process_sort", y=5.328, yerr=0.0 PlotPoint: plot="dbcsr_timings_6cpu_1gpu", name="dbcsr_make_random_matrix", label="dbcsr_make_random_matrix", y=4.465, yerr=0.0 PlotPoint: plot="dbcsr_timings_6cpu_1gpu", name="dbcsr_merge_all", label="dbcsr_merge_all", y=2.73, yerr=0.0 Summary: Performance test works fine. Status: OK Uploading artifacts... done EndDate: 2021-04-20 22:24:34+00:00