imb.5.out (3360B)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 #------------------------------------------------------------ 36 # Intel (R) MPI Benchmarks 4.0, MPI-1 part 37 #------------------------------------------------------------ 38 # Date : Tue Sep 2 20:10:40 2014 39 # Machine : x86_64 40 # System : Linux 41 # Release : 3.10.0-123.4.2.el7.x86_64 42 # Version : #1 SMP Mon Jun 30 16:09:14 UTC 2014 43 # MPI Version : 2.1 44 # MPI Thread Environment: 45 46 # New default behavior from Version 3.2 on: 47 48 # the number of iterations per message size is cut down 49 # dynamically when a certain run time (per message size sample) 50 # is expected to be exceeded. Time limit is defined by variable 51 # "SECS_PER_SAMPLE" (=> IMB_settings.h) 52 # or through the flag => -time 53 54 55 56 # Calling sequence was: 57 58 # /opt/IMB-MPI1 -npmin 25 Exchange 59 60 # Minimum message length in bytes: 0 61 # Maximum message length in bytes: 4194304 62 # 63 # MPI_Datatype : MPI_BYTE 64 # MPI_Datatype for reductions : MPI_FLOAT 65 # MPI_Op : MPI_SUM 66 # 67 # 68 69 # List of Benchmarks to run: 70 71 # Exchange 72 73 #----------------------------------------------------------------------------- 74 # Benchmarking Exchange 75 # #processes = 25 76 #----------------------------------------------------------------------------- 77 #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 78 0 1000 79.13 79.83 79.41 0.00 79 1 1000 77.28 78.16 77.71 0.05 80 2 1000 78.15 78.82 78.51 0.10 81 4 1000 79.77 80.22 80.05 0.19 82 8 1000 80.23 81.02 80.58 0.38 83 16 1000 79.90 80.53 80.27 0.76 84 32 1000 81.08 81.58 81.32 1.50 85 64 1000 85.72 86.31 86.04 2.83 86 128 1000 91.69 92.70 92.19 5.27 87 256 1000 90.80 91.77 91.31 10.64 88 512 1000 98.48 99.63 99.04 19.60 89 1024 1000 103.87 104.86 104.33 37.25 90 2048 1000 138.55 139.84 139.21 55.87 91 4096 1000 258.81 259.67 259.22 60.17 92 8192 1000 365.20 367.03 366.22 85.14 93 16384 1000 477.03 479.55 478.45 130.33 94 32768 1000 1675.90 1686.22 1681.30 74.13 95 65536 640 2755.48 2766.02 2760.60 90.38 96 131072 320 4246.69 4300.53 4270.48 116.26 97 262144 160 12754.71 13106.04 12931.44 76.30 98 524288 80 21575.29 22706.75 22151.51 88.08 99 1048576 40 38364.58 43897.75 41346.74 91.12 100 2097152 20 72575.15 81569.30 76780.63 98.08 101 4194304 10 135199.90 157528.59 146978.85 101.57 102 103 104 # All processes entering MPI_Finalize 105