[mvapich-discuss] Queries reagrding installation of MVAPICH2
Pawan Sood
pawansood at pawansood.com
Fri Feb 5 09:46:57 EST 2010
Skipped content of type multipart/alternative-------------- next part --------------
################################################################################
com Point-to-Point MPI Bandwidth and Latency Benchmark
Version 1.4.0
Run at 02/05/10 14:07:57, with rank 0 on cn28.irsclust.com
################################################################################
Test Processes Op Size (bytes) Ops BW (MB)
-----------------------------------------------------------------
Unidirectional 2 32 100 59.849
Unidirectional 2 64 100 133.678
Unidirectional 2 128 100 223.077
Unidirectional 2 256 100 381.052
Unidirectional 2 512 100 651.796
Unidirectional 2 1024 100 1003.113
Unidirectional 2 2048 100 1110.970
Unidirectional 2 4096 100 1321.899
Unidirectional 2 8192 100 1418.127
Unidirectional 2 16384 100 400.832
Unidirectional 2 32768 100 567.362
Unidirectional 2 65536 100 715.498
Unidirectional 2 131072 100 823.420
Unidirectional 2 262144 100 890.571
Unidirectional 2 524288 100 928.732
Unidirectional 2 1048576 100 948.993
Unidirectional 2 2097152 100 959.365
Unidirectional 2 4194304 100 964.827
Unidirectional 2 8388608 100 967.598
Test Processes Op Size (bytes) Ops BW (MB)
-----------------------------------------------------------------
Bidirectional 2 32 100 12.736
Bidirectional 2 64 100 26.283
Bidirectional 2 128 100 42.596
Bidirectional 2 256 100 81.399
Bidirectional 2 512 100 142.321
Bidirectional 2 1024 100 235.538
Bidirectional 2 2048 100 397.863
Bidirectional 2 4096 100 552.953
Bidirectional 2 8192 100 764.180
Bidirectional 2 16384 100 492.752
Bidirectional 2 32768 100 779.355
Bidirectional 2 65536 100 1089.724
Bidirectional 2 131072 100 1361.221
Bidirectional 2 262144 100 1509.698
Bidirectional 2 524288 100 1801.275
Bidirectional 2 1048576 100 1837.335
Bidirectional 2 2097152 100 1855.337
Bidirectional 2 4194304 100 1893.223
Bidirectional 2 8388608 100 1884.766
Test Processes Op Size (bytes) Ops Latency (us)
-----------------------------------------------------------------
Latency 2 0 100 8.800
Max Unidirectional Bandwidth : 1418.13 for message size of 8192 bytes
Max Bidirectional Bandwidth : 1893.22 for message size of 4194304 bytes
################################################################################
Test Parameters
---------------
Process pair allocation : block
MB size for BW calculation : 1000000
Barrier not included in measurement.
Bandwidth calculated as sum of process bandwidths.
MPI_Wtick returns 0.000010000
MPI_Wtime overhead 0.000001000
-------------- next part --------------
#######################################################################################################
glob MPI Collective Communication Benchmark
Version 1.4.0
Run at 02/05/10 14:09:27, with rank 0 on cn28.irsclust.com
#######################################################################################################
Test Processes Op Size (bytes) Ops Min (us) Avrg (us) Max (us)
Reduce:Double-SUM-R0 2 32 100 1.600 2.000 2.400
Reduce:Double-SUM-R0 2 64 100 1.390 1.690 1.990
Reduce:Double-SUM-R0 2 128 100 1.470 1.835 2.200
Reduce:Double-SUM-R0 2 256 100 1.520 1.865 2.210
Reduce:Double-SUM-R0 2 512 100 1.640 2.045 2.450
Reduce:Double-SUM-R0 2 1024 100 1.670 1.955 2.240
Reduce:Double-SUM-R0 2 2048 100 2.140 2.740 3.340
Reduce:Double-SUM-R0 2 4096 100 3.120 4.295 5.470
Reduce:Double-SUM-R0 2 8192 100 19.240 20.440 21.640
Reduce:Double-SUM-R0 2 16384 100 44.980 44.985 44.990
Reduce:Double-SUM-R0 2 32768 100 70.140 70.185 70.230
Reduce:Double-SUM-R0 2 65536 100 126.700 126.805 126.910
Reduce:Double-SUM-R0 2 131072 100 235.620 235.860 236.100
Reduce:Double-SUM-R0 2 262144 100 453.270 453.760 454.250
Reduce:Double-SUM-R0 2 524288 100 897.610 898.620 899.630
Reduce:Double-SUM-R0 2 1048576 100 1940.550 1942.635 1944.720
Reduce:Double-SUM-R0 2 2097152 100 6379.990 6387.590 6395.190
Reduce:Double-SUM-R0 2 4194304 100 13679.720 13697.195 13714.670
Reduce:Double-SUM-R0 2 8388608 100 28747.580 28782.495 28817.410
Test Parameters
#######################################################################################################
Barrier not included in measurement.
MPI_Wtick returns 0.000010000
MPI_Wtime overhead 0.000001000
#######################################################################################################
-------------- next part --------------
[pawan at cn28 ~]$ /usr/mpi/gcc/mvapich-1.1.0/bin/mpirun_rsh -np 2 -hostfile /gpfs0/pawan/cluster /usr/mpi/gcc/mvapich-1.1.0/tests/presta-1.4.0/globalop
Testing N=4096, 2 tasks
Average work time per task is 0.1813
Average barrier + work time per task is 0.24
Average barrier time per task is 0.02
Average reduce + work time per task is 0.23
Minimum reduce + work time was 0.17
Maximum reduce + work time was 0.29
Average reduce time per task is 0.02
Minimum reduce time was 0.01
Maximum reduce time was 0.03
Average bcast + work time per task is 0.26
Minimum bcast + work time was 0.26
Maximum bcast + work time was 0.26
Average bcast time per task is 0.00
Minimum bcast time was 0.00
Maximum bcast time was 0.00
Average allreduce + work time per task is 0.25
Minimum allreduce + work time was 0.25
Maximum allreduce + work time was 0.25
Average allreduce time per task is 0.02
Minimum allreduce time was 0.02
Maximum allreduce time was 0.02
Average reduce-bcast time per task is 0.04
Minimum reduce-bcast time was 0.04
Maximum reduce-bcast time was 0.04
Average reduce-work-bcast time per task is 0.29
Minimum reduce-work-bcast time was 0.29
Maximum reduce-work-bcast time was 0.29
Average work time per task is 0.1958
Average barrier + work time per task is 0.23
Average barrier time per task is 0.02
Average reduce + work time per task is 0.20
Minimum reduce + work time was 0.15
Maximum reduce + work time was 0.25
Average reduce time per task is 0.02
Minimum reduce time was 0.01
Maximum reduce time was 0.04
Average bcast + work time per task is 0.26
Minimum bcast + work time was 0.26
Maximum bcast + work time was 0.26
Average bcast time per task is 0.00
Minimum bcast time was 0.00
Maximum bcast time was 0.00
Average allreduce + work time per task is 0.24
Minimum allreduce + work time was 0.24
Maximum allreduce + work time was 0.24
Average allreduce time per task is 0.02
Minimum allreduce time was 0.02
Maximum allreduce time was 0.02
Average reduce-bcast time per task is 0.04
Minimum reduce-bcast time was 0.04
Maximum reduce-bcast time was 0.04
Average reduce-work-bcast time per task is 0.29
Minimum reduce-work-bcast time was 0.29
Maximum reduce-work-bcast time was 0.29
Average work time per task is 0.1962
Average barrier + work time per task is 0.24
Average barrier time per task is 0.02
Average reduce + work time per task is 0.20
Minimum reduce + work time was 0.15
Maximum reduce + work time was 0.25
Average reduce time per task is 0.02
Minimum reduce time was 0.01
Maximum reduce time was 0.04
Average bcast + work time per task is 0.23
Minimum bcast + work time was 0.23
Maximum bcast + work time was 0.23
Average bcast time per task is 0.00
Minimum bcast time was 0.00
Maximum bcast time was 0.00
Average allreduce + work time per task is 0.26
Minimum allreduce + work time was 0.26
Maximum allreduce + work time was 0.26
Average allreduce time per task is 0.02
Minimum allreduce time was 0.02
Maximum allreduce time was 0.02
Average reduce-bcast time per task is 0.04
Minimum reduce-bcast time was 0.04
Maximum reduce-bcast time was 0.04
Average reduce-work-bcast time per task is 0.30
Minimum reduce-work-bcast time was 0.30
Maximum reduce-work-bcast time was 0.30
Average elapsed run time was 4.97
All done
-------------- next part --------------
#---------------------------------------------------
# Intel (R) MPI Benchmark Suite V3.1, MPI-1 part
#---------------------------------------------------
# Date : Fri Feb 5 14:13:10 2010
# Machine : x86_64
# System : Linux
# Release : 2.6.9-42.ELsmp
# Version : #1 SMP Wed Jul 12 23:32:02 EDT 2006
# MPI Version : 1.2
# MPI Thread Environment: MPI_THREAD_FUNNELED
# Calling sequence was:
# /usr/mpi/gcc/mvapich-1.1.0/tests/IMB-3.1/IMB-MPI1
# Minimum message length in bytes: 0
# Maximum message length in bytes: 4194304
#
# MPI_Datatype : MPI_BYTE
# MPI_Datatype for reductions : MPI_FLOAT
# MPI_Op : MPI_SUM
#
#
# List of Benchmarks to run:
# PingPong
# PingPing
# Sendrecv
# Exchange
# Allreduce
# Reduce
# Reduce_scatter
# Allgather
# Allgatherv
# Gather
# Gatherv
# Scatter
# Scatterv
# Alltoall
# Alltoallv
# Bcast
# Barrier
#---------------------------------------------------
# Benchmarking PingPong
# #processes = 2
#---------------------------------------------------
#bytes #repetitions t[usec] Mbytes/sec
0 1000 4.10 0.00
1 1000 4.13 0.23
2 1000 4.13 0.46
4 1000 4.20 0.91
8 1000 4.21 1.81
16 1000 4.24 3.60
32 1000 4.40 6.94
64 1000 4.49 13.59
128 1000 5.56 21.94
256 1000 6.00 40.66
512 1000 6.79 71.91
1024 1000 8.11 120.47
2048 1000 9.78 199.64
4096 1000 13.00 300.53
8192 1000 19.52 400.14
16384 1000 38.58 404.99
32768 1000 55.35 564.59
65536 640 89.18 700.86
131072 320 156.80 797.20
262144 160 292.38 855.06
524288 80 564.32 886.02
1048576 40 1109.10 901.63
2097152 20 2203.95 907.46
4194304 10 4419.60 905.06
#---------------------------------------------------
# Benchmarking PingPing
# #processes = 2
#---------------------------------------------------
#bytes #repetitions t[usec] Mbytes/sec
0 1000 4.19 0.00
1 1000 4.20 0.23
2 1000 4.21 0.45
4 1000 4.27 0.89
8 1000 4.32 1.76
16 1000 4.32 3.53
32 1000 4.49 6.80
64 1000 4.59 13.31
128 1000 5.68 21.50
256 1000 6.10 40.00
512 1000 6.90 70.80
1024 1000 8.32 117.42
2048 1000 10.00 195.27
4096 1000 14.04 278.18
8192 1000 20.95 372.91
16384 1000 42.10 371.14
32768 1000 59.10 528.77
65536 640 93.18 670.72
131072 320 161.30 774.97
262144 160 298.04 838.82
524288 80 571.88 874.32
1048576 40 1119.30 893.42
2097152 20 2217.15 902.06
4194304 10 4419.70 905.04
#-----------------------------------------------------------------------------
# Benchmarking Sendrecv
# #processes = 2
#-----------------------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec
0 1000 4.33 4.33 4.33 0.00
1 1000 4.35 4.35 4.35 0.44
2 1000 4.35 4.35 4.35 0.88
4 1000 4.43 4.43 4.43 1.72
8 1000 4.46 4.46 4.46 3.42
16 1000 4.48 4.48 4.48 6.81
32 1000 4.63 4.64 4.64 13.16
64 1000 4.74 4.74 4.74 25.75
128 1000 5.84 5.84 5.84 41.80
256 1000 6.25 6.25 6.25 78.14
512 1000 7.04 7.04 7.04 138.70
1024 1000 8.42 8.42 8.42 231.91
2048 1000 10.32 10.33 10.33 378.29
4096 1000 14.05 14.05 14.05 555.89
8192 1000 20.82 20.83 20.83 749.98
16384 1000 42.59 42.59 42.59 733.65
32768 1000 59.85 59.86 59.86 1044.10
65536 640 94.26 94.27 94.26 1326.04
131072 320 163.45 163.46 163.45 1529.43
262144 160 299.77 299.79 299.78 1667.85
524288 80 578.40 578.42 578.41 1728.83
1048576 40 1120.45 1120.60 1120.53 1784.76
2097152 20 2325.30 2325.35 2325.33 1720.17
4194304 10 4410.30 4410.50 4410.40 1813.85
#-----------------------------------------------------------------------------
# Benchmarking Exchange
# #processes = 2
#-----------------------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec
0 1000 7.52 7.52 7.52 0.00
1 1000 7.55 7.55 7.55 0.50
2 1000 7.54 7.54 7.54 1.01
4 1000 7.64 7.65 7.64 2.00
8 1000 7.67 7.68 7.67 3.98
16 1000 7.69 7.69 7.69 7.94
32 1000 7.86 7.86 7.86 15.53
64 1000 7.83 7.84 7.83 31.16
128 1000 9.36 9.36 9.36 52.19
256 1000 9.91 9.91 9.91 98.50
512 1000 11.01 11.02 11.02 177.27
1024 1000 13.09 13.09 13.09 298.46
2048 1000 16.02 16.02 16.02 487.58
4096 1000 21.44 21.44 21.44 728.68
8192 1000 32.36 32.36 32.36 965.64
16384 1000 75.57 75.57 75.57 827.00
32768 1000 109.76 109.77 109.76 1138.77
65536 640 177.97 177.98 177.98 1404.65
131072 320 314.14 314.16 314.15 1591.55
262144 160 587.05 587.08 587.06 1703.36
524288 80 1133.39 1133.48 1133.43 1764.49
1048576 40 2229.87 2230.00 2229.94 1793.72
2097152 20 4426.75 4427.15 4426.95 1807.03
4194304 10 8910.20 8911.30 8910.75 1795.47
#----------------------------------------------------------------
# Benchmarking Allreduce
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 0.05 0.08 0.07
4 1000 5.23 5.23 5.23
8 1000 5.24 5.24 5.24
16 1000 5.27 5.27 5.27
32 1000 5.45 5.45 5.45
64 1000 5.51 5.52 5.52
128 1000 6.60 6.60 6.60
256 1000 7.09 7.09 7.09
512 1000 8.34 8.34 8.34
1024 1000 10.19 10.19 10.19
2048 1000 12.23 12.23 12.23
4096 1000 16.54 16.54 16.54
8192 1000 25.48 25.48 25.48
16384 1000 58.34 58.34 58.34
32768 1000 93.19 93.19 93.19
65536 640 162.26 162.28 162.27
131072 320 301.63 301.73 301.68
262144 160 581.36 581.88 581.62
524288 80 1266.16 1267.44 1266.80
1048576 40 2851.92 2853.90 2852.91
2097152 20 5681.80 5684.20 5683.00
4194304 10 14605.30 14618.00 14611.65
#----------------------------------------------------------------
# Benchmarking Reduce
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 0.06 0.08 0.07
4 1000 5.67 5.68 5.68
8 1000 5.71 5.71 5.71
16 1000 5.71 5.71 5.71
32 1000 5.88 5.89 5.89
64 1000 5.93 5.94 5.94
128 1000 7.12 7.12 7.12
256 1000 7.51 7.52 7.52
512 1000 8.58 8.59 8.58
1024 1000 9.65 9.67 9.66
2048 1000 11.53 11.55 11.54
4096 1000 15.59 15.61 15.60
8192 1000 24.21 24.23 24.22
16384 1000 52.24 52.25 52.25
32768 1000 81.22 81.25 81.24
65536 640 138.91 138.97 138.94
131072 320 254.37 254.62 254.50
262144 160 517.85 518.87 518.36
524288 80 1153.93 1158.81 1156.37
1048576 40 2358.45 2380.63 2369.54
2097152 20 6082.95 6157.90 6120.43
4194304 10 16046.60 16389.00 16217.80
#----------------------------------------------------------------
# Benchmarking Reduce_scatter
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 0.13 0.14 0.14
4 1000 1.37 4.10 2.73
8 1000 5.76 5.76 5.76
16 1000 5.79 5.79 5.79
32 1000 5.82 5.82 5.82
64 1000 5.97 5.97 5.97
128 1000 6.12 6.13 6.13
256 1000 7.24 7.25 7.25
512 1000 7.85 7.85 7.85
1024 1000 8.89 8.90 8.90
2048 1000 10.93 10.94 10.94
4096 1000 13.04 13.05 13.05
8192 1000 17.99 18.00 17.99
16384 1000 29.28 29.30 29.29
32768 1000 63.30 63.30 63.30
65536 640 99.48 99.48 99.48
131072 320 179.10 179.16 179.13
262144 160 340.52 340.72 340.62
524288 80 514.23 514.89 514.56
1048576 40 1259.40 1263.23 1261.31
2097152 20 2826.40 2848.55 2837.48
4194304 10 6743.30 6756.00 6749.65
#----------------------------------------------------------------
# Benchmarking Allgather
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 0.06 0.07 0.07
1 1000 5.06 5.06 5.06
2 1000 5.02 5.02 5.02
4 1000 5.22 5.22 5.22
8 1000 5.23 5.23 5.23
16 1000 5.27 5.28 5.28
32 1000 5.43 5.43 5.43
64 1000 5.52 5.52 5.52
128 1000 6.58 6.58 6.58
256 1000 7.05 7.05 7.05
512 1000 8.09 8.09 8.09
1024 1000 9.76 9.77 9.76
2048 1000 11.69 11.70 11.70
4096 1000 16.16 16.17 16.17
8192 1000 24.76 24.77 24.77
16384 1000 56.46 56.47 56.47
32768 1000 85.28 85.30 85.29
65536 640 146.73 146.76 146.74
131072 320 279.56 279.65 279.61
262144 160 753.47 754.77 754.12
524288 80 1754.79 1761.80 1758.29
1048576 40 1937.73 1937.98 1937.85
2097152 20 3845.00 3845.45 3845.23
4194304 10 7751.30 7752.40 7751.85
#----------------------------------------------------------------
# Benchmarking Allgatherv
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 0.06 0.07 0.06
1 1000 5.66 5.66 5.66
2 1000 5.67 5.67 5.67
4 1000 5.84 5.84 5.84
8 1000 5.87 5.87 5.87
16 1000 5.91 5.92 5.92
32 1000 6.09 6.09 6.09
64 1000 6.10 6.10 6.10
128 1000 7.14 7.15 7.14
256 1000 7.79 7.79 7.79
512 1000 8.70 8.70 8.70
1024 1000 10.43 10.44 10.44
2048 1000 12.41 12.41 12.41
4096 1000 16.70 16.71 16.70
8192 1000 25.04 25.05 25.05
16384 1000 57.03 57.04 57.04
32768 1000 86.62 86.63 86.62
65536 640 147.53 147.55 147.54
131072 320 268.90 268.99 268.95
262144 160 756.34 757.64 756.99
524288 80 1734.01 1740.95 1737.48
1048576 40 2016.92 2017.15 2017.04
2097152 20 3844.00 3844.35 3844.17
4194304 10 7669.00 7669.30 7669.15
#----------------------------------------------------------------
# Benchmarking Gather
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 0.06 0.08 0.07
1 1000 4.51 4.53 4.52
2 1000 4.46 4.47 4.46
4 1000 4.53 4.53 4.53
8 1000 4.55 4.56 4.56
16 1000 4.57 4.58 4.58
32 1000 4.74 4.74 4.74
64 1000 4.82 4.82 4.82
128 1000 5.90 5.90 5.90
256 1000 6.32 6.32 6.32
512 1000 7.11 7.12 7.12
1024 1000 8.51 8.52 8.52
2048 1000 10.24 10.25 10.24
4096 1000 13.74 13.76 13.75
8192 1000 20.87 20.89 20.88
16384 1000 56.28 56.30 56.29
32768 1000 84.65 84.67 84.66
65536 640 115.32 115.39 115.35
131072 320 209.03 209.29 209.16
262144 160 512.64 514.36 513.50
524288 80 1134.49 1143.68 1139.08
1048576 40 2388.05 2428.80 2408.43
2097152 20 5904.40 6088.65 5996.53
4194304 10 11726.10 12454.90 12090.50
#----------------------------------------------------------------
# Benchmarking Gatherv
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 4.36 4.37 4.37
1 1000 4.35 4.36 4.36
2 1000 4.36 4.36 4.36
4 1000 4.44 4.45 4.45
8 1000 4.48 4.48 4.48
16 1000 4.50 4.50 4.50
32 1000 4.67 4.67 4.67
64 1000 4.78 4.78 4.78
128 1000 5.86 5.87 5.86
256 1000 6.25 6.26 6.25
512 1000 7.02 7.03 7.03
1024 1000 8.42 8.44 8.43
2048 1000 10.12 10.13 10.13
4096 1000 13.55 13.56 13.55
8192 1000 20.53 20.55 20.54
16384 1000 40.07 40.07 40.07
32768 1000 57.97 57.97 57.97
65536 640 95.95 95.97 95.96
131072 320 169.37 169.44 169.40
262144 160 316.57 316.87 316.72
524288 80 691.54 694.79 693.16
1048576 40 1498.03 1518.12 1508.08
2097152 20 2983.20 3063.80 3023.50
4194304 10 6000.20 6322.60 6161.40
#----------------------------------------------------------------
# Benchmarking Scatter
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 0.06 0.09 0.08
1 1000 4.46 4.46 4.46
2 1000 4.43 4.43 4.43
4 1000 4.51 4.51 4.51
8 1000 4.57 4.57 4.57
16 1000 4.55 4.56 4.56
32 1000 4.71 4.72 4.71
64 1000 4.83 4.83 4.83
128 1000 5.88 5.88 5.88
256 1000 6.32 6.32 6.32
512 1000 7.10 7.11 7.11
1024 1000 8.54 8.54 8.54
2048 1000 10.25 10.25 10.25
4096 1000 13.70 13.71 13.70
8192 1000 20.61 20.62 20.61
16384 1000 44.55 44.56 44.56
32768 1000 68.09 68.11 68.10
65536 640 116.29 116.35 116.32
131072 320 212.12 212.32 212.22
262144 160 498.69 499.73 499.21
524288 80 1103.05 1108.21 1105.63
1048576 40 2285.40 2306.00 2295.70
2097152 20 5942.05 6045.25 5993.65
4194304 10 11656.10 12057.70 11856.90
#----------------------------------------------------------------
# Benchmarking Scatterv
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 4.40 4.41 4.40
1 1000 4.40 4.40 4.40
2 1000 4.41 4.41 4.41
4 1000 4.53 4.53 4.53
8 1000 4.55 4.55 4.55
16 1000 4.57 4.57 4.57
32 1000 4.74 4.74 4.74
64 1000 4.80 4.80 4.80
128 1000 5.87 5.88 5.88
256 1000 6.29 6.30 6.29
512 1000 7.08 7.08 7.08
1024 1000 8.47 8.47 8.47
2048 1000 10.15 10.15 10.15
4096 1000 13.53 13.54 13.53
8192 1000 20.52 20.54 20.53
16384 1000 41.74 41.75 41.74
32768 1000 61.51 61.53 61.52
65536 640 102.47 102.53 102.50
131072 320 184.48 184.68 184.58
262144 160 347.94 348.68 348.31
524288 80 685.40 688.52 686.96
1048576 40 1474.60 1493.57 1484.09
2097152 20 2953.00 3032.60 2992.80
4194304 10 5912.20 6231.70 6071.95
#----------------------------------------------------------------
# Benchmarking Alltoall
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 0.06 0.07 0.07
1 1000 4.94 4.94 4.94
2 1000 4.92 4.92 4.92
4 1000 5.01 5.01 5.01
8 1000 5.03 5.04 5.04
16 1000 5.06 5.06 5.06
32 1000 5.23 5.23 5.23
64 1000 5.29 5.29 5.29
128 1000 6.37 6.37 6.37
256 1000 6.82 6.82 6.82
512 1000 7.68 7.69 7.69
1024 1000 9.43 9.43 9.43
2048 1000 11.41 11.41 11.41
4096 1000 15.43 15.44 15.43
8192 1000 23.12 23.12 23.12
16384 1000 45.97 45.98 45.97
32768 1000 65.82 65.82 65.82
65536 640 106.50 106.51 106.51
131072 320 186.43 186.44 186.43
262144 160 361.41 361.44 361.43
524288 80 831.64 831.72 831.68
1048576 40 1973.77 1973.93 1973.85
2097152 20 3840.15 3840.50 3840.32
4194304 10 7667.50 7668.20 7667.85
#----------------------------------------------------------------
# Benchmarking Alltoallv
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 4.80 4.80 4.80
1 1000 4.82 4.82 4.82
2 1000 4.84 4.84 4.84
4 1000 4.94 4.94 4.94
8 1000 4.97 4.97 4.97
16 1000 4.98 4.98 4.98
32 1000 5.16 5.16 5.16
64 1000 5.20 5.20 5.20
128 1000 6.32 6.32 6.32
256 1000 6.74 6.74 6.74
512 1000 7.55 7.56 7.56
1024 1000 9.24 9.25 9.24
2048 1000 11.41 11.42 11.42
4096 1000 15.35 15.35 15.35
8192 1000 22.81 22.82 22.81
16384 1000 45.94 45.95 45.94
32768 1000 65.83 65.83 65.83
65536 640 106.47 106.48 106.47
131072 320 186.39 186.40 186.40
262144 160 361.41 361.43 361.42
524288 80 829.89 829.97 829.93
1048576 40 1972.77 1972.90 1972.84
2097152 20 3838.35 3838.70 3838.53
4194304 10 7667.20 7667.90 7667.55
#----------------------------------------------------------------
# Benchmarking Bcast
# #processes = 2
#----------------------------------------------------------------
#bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
0 1000 0.04 0.07 0.06
1 1000 4.38 4.38 4.38
2 1000 4.36 4.36 4.36
4 1000 4.43 4.43 4.43
8 1000 4.46 4.46 4.46
16 1000 4.48 4.48 4.48
32 1000 4.65 4.65 4.65
64 1000 4.75 4.75 4.75
128 1000 5.81 5.82 5.82
256 1000 6.25 6.25 6.25
512 1000 7.03 7.04 7.04
1024 1000 8.44 8.45 8.44
2048 1000 10.03 10.03 10.03
4096 1000 13.34 13.35 13.35
8192 1000 20.02 20.04 20.03
16384 1000 39.10 39.11 39.10
32768 1000 55.84 55.85 55.85
65536 640 89.62 89.63 89.62
131072 320 157.22 157.24 157.23
262144 160 292.39 292.42 292.41
524288 80 562.62 562.68 562.65
1048576 40 1103.03 1103.22 1103.13
2097152 20 2187.85 2188.10 2187.98
4194304 10 4346.90 4347.40 4347.15
#---------------------------------------------------
# Benchmarking Barrier
# #processes = 2
#---------------------------------------------------
#repetitions t_min[usec] t_max[usec] t_avg[usec]
1000 4.43 4.43 4.43
-------------- next part --------------
# OSU Broadcast Latency Test v3.0
# Size Latency (us)
1 4.28
2 4.27
4 4.31
8 4.32
16 4.34
32 4.54
64 4.60
128 5.67
256 6.20
512 7.11
1024 8.55
2048 10.36
4096 13.83
8192 20.84
16384 38.38
-------------- next part --------------
# OSU MPI Bi-Directional Bandwidth Test v3.0
# Size Bi-Bandwidth (MB/s)
1 2.65
2 5.27
4 9.60
8 19.31
16 36.89
32 70.32
64 141.98
128 264.73
256 471.14
512 768.03
1024 1048.41
2048 1339.73
4096 1332.85
8192 1531.60
16384 1433.71
32768 1643.05
65536 1769.47
131072 1839.06
262144 1874.86
524288 1892.87
1048576 1902.15
2097152 1907.09
4194304 1909.32
-------------- next part --------------
# OSU MPI Latency Test v3.0
# Size Latency (us)
0 4.03
1 4.07
2 4.07
4 4.13
8 4.15
16 4.18
32 4.31
64 4.43
128 5.48
256 5.81
512 6.59
1024 7.83
2048 9.49
4096 12.73
8192 19.30
16384 37.98
32768 55.02
65536 88.82
131072 156.19
262144 291.49
524288 561.89
1048576 1102.08
2097152 2182.45
4194304 4343.50
-------------- next part --------------
# OSU MPI Bandwidth Test v3.0
# Size Bandwidth (MB/s)
1 1.74
2 3.51
4 6.09
8 13.10
16 26.28
32 50.86
64 94.51
128 176.40
256 322.14
512 480.82
1024 645.17
2048 777.23
4096 859.60
8192 905.07
16384 823.70
32768 891.13
65536 929.89
131072 950.52
262144 960.37
524288 965.38
1048576 967.95
2097152 969.25
4194304 969.87
More information about the mvapich-discuss
mailing list