FPMPI-2 |
Features |
|
FPMPI is a simple MPI profiling library. It is intended as a first step towards Applications run which are linked to FPMPI will generate an output file, fpmpi_profile.txt.
on a single processor PC Workstation running Windows 2000, using MPICH. The application is the simple example all_pair.c, which is distributed with FPMPI. |
fpmpi_profile.txt MPI Routine Statistics Explanation of data: Times are the time to perform the operation, e.g., the time for MPI_Send Average times are the average over all processes, e.g., sum (time on each process) / number of processes Min and max values are over all processes (Data is always average/min/max) Amount of data is computed in bytes. For point-to-point operations, it is the data sent or received. For collective operations, it is the data contibuted to the operation. E.g., for an MPI_Bcast, the amount of data is the number of bytes provided by the root, counted only at the root. For synchronizing collective operations, the average, min, and max time spent synchronizing is shown next. Calls by message size shows the fraction of calls that sent messages of a particular size. The bins are 0 bytes, 1-4 bytes, 5-8 bytes, 9-16, 17-32, 33-64, -128, -256, -512, -1024 -4K, -8K, -16K, -32K, -64K, -128K, -256K, -512K, -1M, -4M, -8M, -16M, -32M, -64M, -128M, -256M, -512M, -1GB, >1GB. Each bin is represented by a single digit, representing the 10's of percent of messages within this bin. A 0 represents precisely 0, a . (period) represents more than 0 but less than 10%. A * represents 100%. Messages by message size shows similar information, but for the total message size. The experimental topology information shows the 1-norm distance that the longest point-to-point message travelled, by process. MPI_Pcontrol may be used to control the collection of data Command: F:\Home\Kris\fpmpi-dev\example\all_pair.exe Date: Fri Jul 30 04:01:38 2004 Processes: 2 Execute time: 0.078 Timing Stats: [seconds] [min/max] [min rank/max rank] wall-clock: 0.078 sec 0.031000 / 0.125000 1 / 0 Routine Calls ave time min time max time ave len min len max len ave sync min sync max sync Calls by message size % data by message size Routine Calls ave time min time max time ave len min len max len Calls by message size % data by message size MPI_Send 6 0.000014 0.000000 0.000015 4532.000000 0.000000 7464.000000 20000000008000000000000000000 0000000000*000000000000000000 MPI_Sendrecv 3 0.000071 0.000000 0.000087 800.000000 0.000000 1600.000000 60000000003000000000000000000 0000000000*000000000000000000 MPI_Sendrecv_replace 1 0.000032 0.000000 0.000065 1332.000000 0.000000 2664.000000 00000000009000000000000000000 0000000000*000000000000000000 MPI_Isend 2 0.000006 0.000000 0.000009 1600.000000 0.000000 1600.000000 0000000000*000000000000000000 0000000000*000000000000000000 MPI_Issend 2 0.000007 0.000000 0.000008 2664.000000 0.000000 2664.000000 0000000000*000000000000000000 0000000000*000000000000000000 MPI_Irsend 2 0.000009 0.000000 0.000009 2664.000000 0.000000 2664.000000 0000000000*000000000000000000 0000000000*000000000000000000 MPI_Rsend 1 0.000002 0.000000 0.000005 800.000000 0.000000 1600.000000 00000000009000000000000000000 0000000000*000000000000000000 MPI_Ssend 2 0.000052 0.000000 0.000054 2664.000000 0.000000 2664.000000 0000000000*000000000000000000 0000000000*000000000000000000 MPI_Recv 10 0.000313 0.000000 0.000439 9328.000000 0.000000 12792.000000 10000000009000000000000000000 0000000000*000000000000000000 MPI_Irecv 5 0.000005 0.000000 0.000006 20000.000000 0.000000 24000.000000 00000000000*00000000000000000 00000000000*00000000000000000 MPI_Waitall 3 0.000052 0.000002 0.000102 MPI_Waitany 2 0.000053 0.000000 0.000106 MPI_Waitsome 2 0.000035 0.000000 0.000070 MPI_Testall 4 0.000070 0.000000 0.000140 MPI_Testsome 4 0.000076 0.000000 0.000151 MPI_Iprobe 5 0.000030 0.000000 0.000061 MPI_Probe 1 0.000023 0.000000 0.000046 MPI_Comm_dup 2 MPI_Comm_dup(MPI_COMM_WORLD) 2 1-norm distance of point-to-point with an assumed 2-d topology 1 1 Detailed partner data: source dest1 dest2 ... 0 1 1 0 |