1001114.edique02::Benchmark_Time
powered by IPM
back
command: ./mini_dft+ipm -in MDFT_10.in -ntg 9 -ndiag 625
region: Benchmark_Time
state: unknown
username: baustin
group:
host: nid04446 (x86_64_Linux)
mpi_tasks: 648 on 54 hosts
wallclock: 1.69696e+03 sec
#exits: 648/648
%comm: 19.9931492090038
gflop/sec: 0

Computation

Event Count Pop

Communication

% of MPI Time
HPM Counter Statistics
Event Ntasks Avg Min(rank) Max(rank)
Communication Event Statistics (0.00% detail, 2.1985e+05 error)
  Buffer Size Ncalls Total Time Min Time Max Time %MPI %Wall
MPI_Alltoallv 32768 1152144 41910.313 4.054e-03 3.559e-01 19.06 3.81
MPI_Barrier 0 5908395 17920.664 1.383e-05 2.169e+01 8.15 1.63
MPI_Alltoallv 262144 1152144 8981.953 1.424e-03 1.742e-01 4.09 0.82
MPI_Bcast 5120 2406225 4220.813 9.537e-07 4.385e-02 1.92 0.38
MPI_Bcast 2560 127850 2371.516 0.000e+00 1.184e+00 1.08 0.22
MPI_Reduce 786432 843000 1412.959 1.919e-04 6.373e-02 0.64 0.13
MPI_Bcast 786432 405000 1234.547 5.400e-04 2.592e-02 0.56 0.11
MPI_Allreduce 786432 207360 1163.814 2.278e-03 3.059e-02 0.53 0.11
MPI_Recv 5120 7203930 1455.589 0.000e+00 1.767e-01 0.66 0.13
MPI_Reduce 32768 421200 1656.816 6.390e-05 6.360e-02 0.75 0.15
MPI_Recv 4 15000 201.750 0.000e+00 5.657e-02 0.09 0.02
MPI_Bcast 1572864 15000 544.480 1.186e-03 8.876e-02 0.25 0.05
MPI_Allreduce 8 20809 204.848 1.693e-05 8.284e-02 0.09 0.02
MPI_Bcast 32768 195025 402.901 2.909e-05 2.252e-01 0.18 0.04
MPI_Recv 1572864 51409 388.489 1.509e-04 7.949e-02 0.18 0.04
MPI_Recv 81920 698583 324.218 5.960e-06 1.838e-01 0.15 0.03
MPI_Bcast 163840 74250 300.414 2.289e-05 5.823e-01 0.14 0.03
MPI_Reduce 163840 74250 254.315 3.290e-05 1.083e+00 0.12 0.02
MPI_Send 81920 698040 229.794 5.960e-06 8.526e-01 0.10 0.02
MPI_Sendrecv_replace 1572864 93750 211.601 2.668e-04 2.066e-02 0.10 0.02
MPI_Alltoallv 0 2304 202.466 5.200e-02 1.445e-01 0.09 0.02
MPI_Reduce 1572864 15000 172.762 3.252e-04 5.943e-02 0.08 0.02
MPI_Bcast 1024 1355016 120.763 0.000e+00 9.587e-03 0.05 0.01
MPI_Alltoall 4 648 225.681 8.049e-03 5.723e-01 0.10 0.02
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
cumulative values, values
Message Buffer Size Distributions: Ncalls
cumulative values, values
Communication Topology : point to point data flow
data sent , data recv , time spent , map_data file map_adjacency file
back