|
|
|
|
HPM Counter Statistics | ||||
---|---|---|---|---|
Event | Ntasks | Avg | Min(rank) | Max(rank) |
PAPI_FP_OPS | * | 3020341686.77 | 2962849966 (10706) | 3108475920 (7036) |
PAPI_L1_DCA | * | 403168018215.03 | 393571844381 (4172) | 410686575226 (1025) |
PAPI_L1_DCM | * | 1067312225.34 | 1003898306 (8666) | 1487111167 (6978) |
PAPI_TOT_INS | * | 884441767556.03 | 867107658475 (0) | 898764952737 (1025) |
Communication Event Statistics (100.00% detail, -2.1744e-02 error) | |||||||
---|---|---|---|---|---|---|---|
Buffer Size | Ncalls | Total Time | Min Time | Max Time | %MPI | %Wall | |
MPI_Alltoallv | 0 | 1492992 | 328067.943 | 1.838e-01 | 2.988e-01 | 41.65 | 8.38 |
MPI_Allgather | 8 | 843264 | 106565.044 | 7.272e-04 | 1.618e+00 | 13.53 | 2.72 |
MPI_Waitsome | 0 | 32399453 | 62615.635 | 0.000e+00 | 6.420e-01 | 7.95 | 1.60 |
MPI_Send | 8192 | 254890632 | 57793.603 | 3.099e-06 | 6.292e-01 | 7.34 | 1.48 |
MPI_Alltoall | 4 | 1492992 | 49093.091 | 9.957e-03 | 1.880e-01 | 6.23 | 1.25 |
MPI_Waitall | 64 | 2197698 | 40427.160 | 0.000e+00 | 1.698e-01 | 5.13 | 1.03 |
MPI_Allreduce | 8 | 8921016 | 27549.422 | 5.317e-05 | 1.588e-01 | 3.50 | 0.70 |
MPI_Waitall | 3072 | 25064208 | 13643.390 | 0.000e+00 | 5.806e-02 | 1.73 | 0.35 |
MPI_Send | 16384 | 3597776 | 12340.313 | 1.001e-05 | 4.953e-01 | 1.57 | 0.32 |
MPI_Send | 32768 | 3842504 | 11135.663 | 4.888e-05 | 1.069e-01 | 1.41 | 0.28 |
MPI_Recv | 4 | 13792 | 9055.653 | 1.542e+00 | 1.614e+00 | 1.15 | 0.23 |
MPI_Waitall | 49152 | 32455104 | 8352.810 | 0.000e+00 | 3.463e-02 | 1.06 | 0.21 |
MPI_Waitall | 12288 | 16930320 | 7642.978 | 0.000e+00 | 3.073e-02 | 0.97 | 0.20 |
MPI_Allgather | 4 | 262656 | 6111.589 | 8.234e-03 | 9.017e-02 | 0.78 | 0.16 |
MPI_Waitall | 768 | 15328872 | 5484.922 | 0.000e+00 | 3.491e-02 | 0.70 | 0.14 |
MPI_Allreduce | 24 | 262656 | 4893.147 | 1.319e-02 | 6.188e-02 | 0.62 | 0.12 |
MPI_Send | 327680 | 666944 | 3670.110 | 2.718e-04 | 1.815e-01 | 0.47 | 0.09 |
MPI_Send | 98304 | 1784472 | 2955.851 | 4.602e-05 | 1.376e-01 | 0.38 | 0.08 |
MPI_Waitall | 40960 | 8851392 | 2341.271 | 0.000e+00 | 2.820e-02 | 0.30 | 0.06 |
MPI_Send | 12288 | 3228832 | 2124.007 | 1.311e-05 | 1.151e-01 | 0.27 | 0.05 |
MPI_Waitall | 10240 | 4617360 | 2096.469 | 0.000e+00 | 2.741e-02 | 0.27 | 0.05 |
MPI_Send | 40960 | 1073352 | 1994.923 | 1.400e-04 | 5.762e-02 | 0.25 | 0.05 |
MPI_Waitall | 448 | 1705752 | 1740.729 | 0.000e+00 | 3.542e-02 | 0.22 | 0.04 |
MPI_Waitall | 3584 | 1693350 | 1719.490 | 2.146e-06 | 3.960e-02 | 0.22 | 0.04 |
MPI_Waitall | 1024 | 1703208 | 1649.596 | 1.907e-06 | 3.324e-02 | 0.21 | 0.04 |
MPI_Waitall | 2560 | 4303752 | 1561.271 | 0.000e+00 | 6.021e-02 | 0.20 | 0.04 |
MPI_Waitall | 640 | 4342608 | 1543.041 | 0.000e+00 | 3.177e-02 | 0.20 | 0.04 |
MPI_Waitall | 32768 | 1481656 | 1517.730 | 0.000e+00 | 2.857e-02 | 0.19 | 0.04 |
MPI_Allreduce | 4 | 193536 | 1261.347 | 5.550e-04 | 4.398e-02 | 0.16 | 0.03 |
MPI_Send | 20480 | 507432 | 772.138 | 6.294e-05 | 9.447e-02 | 0.10 | 0.02 |
MPI_Send | 2048 | 132573648 | 762.728 | 0.000e+00 | 9.266e-03 | 0.10 | 0.02 |
MPI_Send | 81920 | 957352 | 704.988 | 4.411e-05 | 5.007e-02 | 0.09 | 0.02 |
MPI_Send | 8 | 106998576 | 659.482 | 0.000e+00 | 4.267e-02 | 0.08 | 0.02 |
MPI_Send | 256 | 95400152 | 596.058 | 0.000e+00 | 5.177e-02 | 0.08 | 0.02 |
MPI_Bcast | 8 | 82944 | 519.831 | 1.352e-03 | 1.423e-02 | 0.07 | 0.01 |
MPI_Send | 512 | 141391392 | 461.811 | 0.000e+00 | 5.573e-02 | 0.06 | 0.01 |
MPI_Waitall | 896 | 477000 | 421.257 | 0.000e+00 | 3.358e-02 | 0.05 | 0.01 |
MPI_Send | 64 | 30269716 | 417.681 | 0.000e+00 | 7.170e-02 | 0.05 | 0.01 |
MPI_Waitall | 320 | 477000 | 408.943 | 1.907e-06 | 2.793e-02 | 0.05 | 0.01 |
MPI_Send | 128 | 152355424 | 404.664 | 0.000e+00 | 1.105e-02 | 0.05 | 0.01 |
MPI_Gatherv | 0 | 60480 | 400.061 | 2.066e-03 | 1.433e-02 | 0.05 | 0.01 |
Load balance by task: HPM counters |
---|
by MPI rank, by MPI time |
Load balance by task: memory, flops, timings |
by MPI rank, by MPI time |
Communication balance by task (sorted by MPI time) |
by MPI rank , time detail by MPI time , time detail by rank , call list |
Message Buffer Size Distributions: time |
|
Message Buffer Size Distributions: Ncalls |
|
Communication Topology : point to point data flow |
|
Switch Traffic (volume by node) |
|
Memory usage by node |
|