Open-MX 0.5.0 (--disable-mx-wire --disable-endian OMX_RCACHE=1 dmaengine=1) MPICH-MX 1.2.7..5 Dual quad-core Xeon (E5345, 2.33GHz, C1E sleeping disabled) Local communication only, through Open-MX shared-memory path Processes on cores #0 and #2 (closest) #--------------------------------------------------- # Intel (R) MPI Benchmark Suite V2.3, MPI-1 part #--------------------------------------------------- # Date : Tue Apr 29 21:35:03 2008 # Machine : x86_64# System : Linux # Release : 2.6.23-1-amd64 # Version : #1 SMP Fri Nov 30 12:19:15 UTC 2007 # # 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 # Alltoall # Bcast # Barrier #--------------------------------------------------- # Benchmarking PingPong # #processes = 2 #--------------------------------------------------- #bytes #repetitions t[usec] Mbytes/sec 0 1000 0.70 0.00 1 1000 0.71 1.35 2 1000 0.72 2.65 4 1000 0.75 5.11 8 1000 0.70 10.91 16 1000 0.72 21.34 32 1000 0.73 41.72 64 1000 0.99 61.78 128 1000 1.01 121.04 256 1000 1.30 187.22 512 1000 1.48 330.94 1024 1000 1.88 518.35 2048 1000 2.70 722.05 4096 1000 4.18 935.53 8192 1000 6.95 1123.38 16384 1000 12.05 1296.42 32768 1000 22.03 1418.48 65536 640 12.63 4949.91 131072 320 23.14 5401.72 262144 160 44.66 5597.54 524288 80 90.30 5537.03 1048576 40 329.81 3032.02 2097152 20 837.13 2389.12 4194304 10 1673.45 2390.27 #--------------------------------------------------- # Benchmarking PingPing # #processes = 2 #--------------------------------------------------- #bytes #repetitions t[usec] Mbytes/sec 0 1000 1.22 0.00 1 1000 1.35 0.71 2 1000 1.35 1.41 4 1000 1.39 2.75 8 1000 1.33 5.72 16 1000 1.34 11.37 32 1000 1.39 21.89 64 1000 1.51 40.29 128 1000 1.56 78.00 256 1000 2.22 109.88 512 1000 2.44 200.12 1024 1000 2.97 328.47 2048 1000 4.09 478.03 4096 1000 6.21 629.52 8192 1000 11.59 673.91 16384 1000 24.16 646.81 32768 1000 43.76 714.14 65536 640 15.13 4132.22 131072 320 27.25 4586.57 262144 160 52.44 4767.61 524288 80 108.34 4615.21 1048576 40 491.58 2034.27 2097152 20 1320.30 1514.80 4194304 10 2624.01 1524.38 #----------------------------------------------------------------------------- # Benchmarking Sendrecv # #processes = 2 #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 1.35 1.35 1.35 0.00 1 1000 1.38 1.38 1.38 1.38 2 1000 1.36 1.36 1.36 2.80 4 1000 1.41 1.41 1.41 5.40 8 1000 1.36 1.36 1.36 11.24 16 1000 1.35 1.35 1.35 22.54 32 1000 1.36 1.36 1.36 44.78 64 1000 1.52 1.52 1.52 80.20 128 1000 1.65 1.65 1.65 148.32 256 1000 2.29 2.29 2.29 212.87 512 1000 2.52 2.52 2.52 387.22 1024 1000 3.02 3.02 3.02 647.18 2048 1000 4.10 4.10 4.10 953.00 4096 1000 6.13 6.14 6.14 1273.24 8192 1000 11.20 11.20 11.20 1394.71 16384 1000 22.33 22.33 22.33 1399.16 32768 1000 43.05 43.06 43.05 1451.63 65536 640 15.59 15.59 15.59 8016.06 131072 320 27.91 27.92 27.91 8955.49 262144 160 53.29 53.31 53.30 9379.82 524288 80 112.02 112.02 112.02 8927.38 1048576 40 491.88 491.88 491.88 4066.07 2097152 20 1338.10 1338.10 1338.10 2989.31 4194304 10 2622.01 2622.01 2622.01 3051.10 #----------------------------------------------------------------------------- # Benchmarking Exchange # #processes = 2 #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 2.44 2.44 2.44 0.00 1 1000 2.60 2.60 2.60 1.47 2 1000 2.57 2.57 2.57 2.97 4 1000 2.65 2.65 2.65 5.76 8 1000 2.59 2.60 2.60 11.76 16 1000 2.59 2.59 2.59 23.55 32 1000 2.67 2.67 2.67 45.69 64 1000 2.91 2.91 2.91 83.81 128 1000 3.22 3.22 3.22 151.78 256 1000 4.38 4.38 4.38 222.71 512 1000 4.91 4.91 4.91 397.79 1024 1000 5.90 5.90 5.90 662.17 2048 1000 8.17 8.17 8.17 956.00 4096 1000 12.65 12.65 12.65 1235.06 8192 1000 23.04 23.05 23.05 1355.98 16384 1000 47.95 47.95 47.95 1303.52 32768 1000 87.76 87.76 87.76 1424.29 65536 640 30.07 30.08 30.07 8312.55 131072 320 53.90 53.90 53.90 9275.84 262144 160 102.14 102.14 102.14 9790.77 524288 80 205.69 205.70 205.69 9722.82 1048576 40 935.25 935.27 935.26 4276.82 2097152 20 2619.90 2619.95 2619.92 3053.50 4194304 10 5216.31 5216.31 5216.31 3067.30 #---------------------------------------------------------------- # Benchmarking Allreduce # #processes = 2 #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.03 0.03 0.03 4 1000 2.22 2.22 2.22 8 1000 2.26 2.26 2.26 16 1000 2.28 2.28 2.28 32 1000 2.27 2.27 2.27 64 1000 2.42 2.42 2.42 128 1000 2.63 2.63 2.63 256 1000 3.35 3.35 3.35 512 1000 3.65 3.65 3.65 1024 1000 4.36 4.36 4.36 2048 1000 5.77 5.77 5.77 4096 1000 10.06 10.06 10.06 8192 1000 15.43 15.43 15.43 16384 1000 27.67 27.68 27.68 32768 1000 55.15 55.16 55.15 65536 640 106.94 106.95 106.95 131072 320 77.35 77.35 77.35 262144 160 231.78 231.78 231.78 524288 80 478.86 478.86 478.86 1048576 40 1710.83 1710.85 1710.84 2097152 20 5845.30 5845.30 5845.30 4194304 10 10676.69 10676.69 10676.69 #---------------------------------------------------------------- # Benchmarking Reduce # #processes = 2 #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.03 0.03 0.03 4 1000 1.59 1.59 1.59 8 1000 1.57 1.57 1.57 16 1000 1.58 1.58 1.58 32 1000 1.72 1.72 1.72 64 1000 2.18 2.18 2.18 128 1000 2.23 2.23 2.23 256 1000 2.39 2.39 2.39 512 1000 2.66 2.66 2.66 1024 1000 3.39 3.39 3.39 2048 1000 4.55 4.55 4.55 4096 1000 9.29 9.29 9.29 8192 1000 14.26 14.26 14.26 16384 1000 24.56 24.56 24.56 32768 1000 45.42 45.42 45.42 65536 640 85.73 85.74 85.74 131072 320 118.91 118.91 118.91 262144 160 229.00 229.00 229.00 524288 80 631.98 631.99 631.98 1048576 40 2511.43 2511.43 2511.43 2097152 20 6099.65 6099.65 6099.65 4194304 10 11352.80 11386.61 11369.71 #---------------------------------------------------------------- # Benchmarking Reduce_scatter # #processes = 2 #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.09 0.10 0.10 4 1000 2.02 4.14 3.08 8 1000 3.32 3.32 3.32 16 1000 3.26 3.26 3.26 32 1000 3.10 3.10 3.10 64 1000 3.17 3.17 3.17 128 1000 3.39 3.39 3.39 256 1000 3.69 3.69 3.69 512 1000 4.47 4.47 4.47 1024 1000 5.07 5.07 5.07 2048 1000 5.74 5.74 5.74 4096 1000 7.60 7.60 7.60 8192 1000 11.43 11.43 11.43 16384 1000 19.91 19.91 19.91 32768 1000 38.21 38.21 38.21 65536 640 70.50 70.51 70.50 131072 320 92.08 92.09 92.08 262144 160 176.64 176.67 176.66 524288 80 369.41 369.41 369.41 1048576 40 981.23 981.32 981.28 2097152 20 3526.79 3529.24 3528.02 4194304 10 6603.50 6607.41 6605.46 #---------------------------------------------------------------- # Benchmarking Allgather # #processes = 2 #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.03 0.03 0.03 1 1000 2.13 2.13 2.13 2 1000 2.12 2.12 2.12 4 1000 2.14 2.14 2.14 8 1000 2.12 2.12 2.12 16 1000 2.11 2.11 2.11 32 1000 2.17 2.17 2.17 64 1000 2.22 2.22 2.22 128 1000 2.47 2.47 2.47 256 1000 3.11 3.11 3.11 512 1000 3.39 3.39 3.39 1024 1000 3.87 3.87 3.87 2048 1000 4.92 4.92 4.92 4096 1000 7.38 7.38 7.38 8192 1000 12.99 12.99 12.99 16384 1000 25.74 25.74 25.74 32768 1000 49.15 49.15 49.15 65536 640 26.23 26.23 26.23 131072 320 48.22 48.22 48.22 262144 160 96.82 96.83 96.83 524288 80 517.27 517.27 517.27 1048576 40 2348.73 2348.73 2348.73 2097152 20 3341.40 3341.40 3341.40 4194304 10 7555.82 7555.82 7555.82 #---------------------------------------------------------------- # Benchmarking Allgatherv # #processes = 2 #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.04 0.04 0.04 1 1000 3.57 3.57 3.57 2 1000 3.55 3.55 3.55 4 1000 3.60 3.60 3.60 8 1000 3.56 3.56 3.56 16 1000 3.62 3.62 3.62 32 1000 3.62 3.62 3.62 64 1000 3.66 3.66 3.66 128 1000 3.88 3.89 3.88 256 1000 4.64 4.64 4.64 512 1000 4.88 4.88 4.88 1024 1000 5.44 5.44 5.44 2048 1000 6.73 6.73 6.73 4096 1000 9.54 9.54 9.54 8192 1000 15.95 15.96 15.95 16384 1000 32.10 32.10 32.10 32768 1000 58.06 58.07 58.07 65536 640 66.49 66.50 66.50 131072 320 124.35 124.38 124.37 262144 160 95.24 95.24 95.24 524288 80 518.72 518.72 518.72 1048576 40 2351.02 2351.02 2351.02 2097152 20 3294.85 3294.85 3294.85 4194304 10 7552.60 7552.70 7552.65 #---------------------------------------------------------------- # Benchmarking Alltoall # #processes = 2 #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.03 0.03 0.03 1 1000 2.32 2.32 2.32 2 1000 2.28 2.28 2.28 4 1000 2.30 2.30 2.30 8 1000 2.25 2.25 2.25 16 1000 2.26 2.26 2.26 32 1000 2.28 2.28 2.28 64 1000 2.28 2.28 2.28 128 1000 2.50 2.50 2.50 256 1000 3.11 3.11 3.11 512 1000 3.46 3.46 3.46 1024 1000 3.86 3.86 3.86 2048 1000 5.07 5.07 5.07 4096 1000 7.66 7.66 7.66 8192 1000 13.53 13.53 13.53 16384 1000 26.08 26.08 26.08 32768 1000 48.61 48.61 48.61 65536 640 26.28 26.28 26.28 131072 320 48.06 48.06 48.06 262144 160 108.36 108.36 108.36 524288 80 435.96 435.98 435.97 1048576 40 2334.15 2334.15 2334.15 2097152 20 3416.94 3416.94 3416.94 4194304 10 7343.20 7343.29 7343.24 #---------------------------------------------------------------- # Benchmarking Bcast # #processes = 2 #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.02 0.02 0.02 1 1000 0.79 0.79 0.79 2 1000 0.79 0.79 0.79 4 1000 0.82 0.82 0.82 8 1000 0.78 0.78 0.78 16 1000 0.78 0.78 0.78 32 1000 0.80 0.80 0.80 64 1000 1.05 1.05 1.05 128 1000 1.11 1.11 1.11 256 1000 1.37 1.37 1.37 512 1000 1.59 1.59 1.59 1024 1000 1.99 1.99 1.99 2048 1000 2.88 2.88 2.88 4096 1000 4.40 4.40 4.40 8192 1000 7.36 7.36 7.36 16384 1000 12.44 12.44 12.44 32768 1000 22.46 22.46 22.46 65536 640 13.10 13.10 13.10 131072 320 23.74 23.75 23.75 262144 160 46.09 46.09 46.09 524288 80 96.71 96.71 96.71 1048576 40 332.90 332.90 332.90 2097152 20 839.10 839.10 839.10 4194304 10 1673.41 1673.41 1673.41 #--------------------------------------------------- # Benchmarking Barrier # #processes = 2 #--------------------------------------------------- #repetitions t_min[usec] t_max[usec] t_avg[usec] 1000 1.41 1.41 1.41