imb.3.out (3340B)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 #------------------------------------------------------------ 18 # Intel (R) MPI Benchmarks 4.0, MPI-1 part 19 #------------------------------------------------------------ 20 # Date : Tue Sep 2 20:08:36 2014 21 # Machine : x86_64 22 # System : Linux 23 # Release : 3.10.0-123.4.2.el7.x86_64 24 # Version : #1 SMP Mon Jun 30 16:09:14 UTC 2014 25 # MPI Version : 2.1 26 # MPI Thread Environment: 27 28 # New default behavior from Version 3.2 on: 29 30 # the number of iterations per message size is cut down 31 # dynamically when a certain run time (per message size sample) 32 # is expected to be exceeded. Time limit is defined by variable 33 # "SECS_PER_SAMPLE" (=> IMB_settings.h) 34 # or through the flag => -time 35 36 37 38 # Calling sequence was: 39 40 # /opt/IMB-MPI1 -npmin 9 Exchange 41 42 # Minimum message length in bytes: 0 43 # Maximum message length in bytes: 4194304 44 # 45 # MPI_Datatype : MPI_BYTE 46 # MPI_Datatype for reductions : MPI_FLOAT 47 # MPI_Op : MPI_SUM 48 # 49 # 50 51 # List of Benchmarks to run: 52 53 # Exchange 54 55 #----------------------------------------------------------------------------- 56 # Benchmarking Exchange 57 # #processes = 9 58 #----------------------------------------------------------------------------- 59 #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 60 0 1000 83.47 83.72 83.60 0.00 61 1 1000 80.98 81.23 81.11 0.05 62 2 1000 77.56 77.78 77.68 0.10 63 4 1000 78.04 78.16 78.10 0.20 64 8 1000 78.34 78.64 78.49 0.39 65 16 1000 77.51 77.78 77.67 0.78 66 32 1000 77.48 78.15 77.64 1.56 67 64 1000 77.90 78.23 78.10 3.12 68 128 1000 83.74 84.07 83.96 5.81 69 256 1000 88.70 88.94 88.79 10.98 70 512 1000 96.55 97.22 96.79 20.09 71 1024 1000 102.43 102.78 102.58 38.01 72 2048 1000 126.63 126.85 126.70 61.59 73 4096 1000 258.96 259.28 259.14 60.26 74 8192 1000 348.34 348.72 348.54 89.61 75 16384 1000 492.73 493.38 493.03 126.68 76 32768 1000 1257.23 1260.81 1259.21 99.14 77 65536 640 2748.78 2751.00 2750.14 90.88 78 131072 320 5145.39 5154.34 5148.53 97.01 79 262144 160 14654.73 14767.01 14712.56 67.72 80 524288 80 20009.40 20418.24 20188.72 97.95 81 1048576 40 39012.20 40482.40 39862.64 98.81 82 2097152 20 76761.84 83431.35 80447.34 95.89 83 4194304 10 140151.60 161396.50 152266.14 99.13 84 85 86 # All processes entering MPI_Finalize 87