#--------------------------------------------------- # PALLAS MPI Benchmark Suite V2.2, MPI-1 part #--------------------------------------------------- # Date : Fri Apr 16 19:54:31 2004 # Machine : i686# System : Linux # Release : 2.4.22-1.2174.nptlsmp # Version : #1 SMP Wed Feb 18 16:22:03 EST 2004 # # 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 ) # ( 6 additional processes waiting in MPI_Barrier) #--------------------------------------------------- #bytes #repetitions t[usec] Mbytes/sec 0 1000 78.48 0.00 1 1000 80.75 0.01 2 1000 85.03 0.02 4 1000 85.35 0.04 8 1000 83.86 0.09 16 1000 83.51 0.18 32 1000 86.50 0.35 64 1000 100.51 0.61 128 1000 116.18 1.05 256 1000 134.29 1.82 512 1000 184.92 2.64 1024 1000 276.07 3.54 2048 1000 392.60 4.97 4096 1000 571.47 6.84 8192 1000 951.94 8.21 16384 1000 1682.06 9.29 32768 1000 3158.71 9.89 65536 640 6201.56 10.08 131072 320 12428.24 10.06 262144 160 24831.56 10.07 524288 80 49183.38 10.17 1048576 40 97855.02 10.22 2097152 20 195101.82 10.25 4194304 10 388291.60 10.30 #--------------------------------------------------- # Benchmarking PingPing # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #--------------------------------------------------- #bytes #repetitions t[usec] Mbytes/sec 0 1000 85.65 0.00 1 1000 86.69 0.01 2 1000 86.88 0.02 4 1000 90.78 0.04 8 1000 91.19 0.08 16 1000 93.74 0.16 32 1000 96.21 0.32 64 1000 104.71 0.58 128 1000 118.91 1.03 256 1000 142.86 1.71 512 1000 185.39 2.63 1024 1000 276.64 3.53 2048 1000 399.61 4.89 4096 1000 580.28 6.73 8192 1000 976.60 8.00 16384 1000 1736.43 9.00 32768 1000 3313.64 9.43 65536 640 6548.08 9.54 131072 320 16893.52 7.40 262144 160 40750.08 6.13 524288 80 89080.24 5.61 1048576 40 183769.95 5.44 2097152 20 374716.05 5.34 4194304 10 757801.50 5.28 #----------------------------------------------------------------------------- # Benchmarking Sendrecv # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 89.06 89.23 89.15 0.00 1 1000 90.89 90.99 90.94 0.02 2 1000 88.87 88.92 88.90 0.04 4 1000 90.90 90.95 90.93 0.08 8 1000 90.39 90.50 90.44 0.17 16 1000 91.23 91.31 91.27 0.33 32 1000 97.06 97.08 97.07 0.63 64 1000 105.35 105.47 105.41 1.16 128 1000 121.49 121.56 121.53 2.01 256 1000 140.83 140.94 140.88 3.46 512 1000 187.09 187.17 187.13 5.22 1024 1000 278.26 278.38 278.32 7.02 2048 1000 401.15 401.32 401.24 9.73 4096 1000 578.48 578.73 578.61 13.50 8192 1000 976.77 976.92 976.85 15.99 16384 1000 1737.12 1737.26 1737.19 17.99 32768 1000 3314.77 3314.94 3314.85 18.85 65536 640 6552.79 6553.00 6552.90 19.08 131072 320 16805.92 16815.22 16810.57 14.87 262144 160 40680.77 40732.29 40706.53 12.28 524288 80 88203.14 88290.25 88246.69 11.33 1048576 40 184693.52 184876.60 184785.06 10.82 2097152 20 374517.35 374870.85 374694.10 10.67 4194304 10 761393.10 762169.80 761781.45 10.50 #----------------------------------------------------------------------------- # Benchmarking Sendrecv # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 536.24 536.36 536.32 0.00 1 1000 655.22 655.31 655.26 0.00 2 1000 1461.34 1461.59 1461.45 0.00 4 1000 679.96 680.18 680.04 0.01 8 1000 1732.81 1765.81 1757.52 0.01 16 1000 1111.43 1111.64 1111.53 0.03 32 1000 468.87 469.08 468.92 0.13 64 1000 101.68 101.81 101.77 1.20 128 1000 113.47 113.58 113.53 2.15 256 1000 135.60 135.71 135.65 3.60 512 1000 187.85 188.01 187.95 5.19 1024 1000 277.47 277.73 277.56 7.03 2048 1000 385.13 385.40 385.24 10.14 4096 1000 577.39 577.79 577.62 13.52 8192 1000 947.29 947.71 947.46 16.49 16384 1000 1691.08 1692.93 1692.39 18.46 32768 1000 3279.94 3280.60 3280.31 19.05 65536 640 6934.30 6939.21 6936.84 18.01 131072 320 23780.68 23836.84 23808.86 10.49 262144 160 47398.04 47588.86 47478.19 10.51 524288 80 93747.10 94385.94 94040.20 10.59 1048576 40 187291.52 189683.73 188415.23 10.54 2097152 20 374894.10 384299.30 379483.59 10.41 4194304 10 857843.90 897053.00 877801.20 8.92 #----------------------------------------------------------------------------- # Benchmarking Sendrecv # ( #processes = 8 ) #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 1658.45 1696.09 1672.64 0.00 1 1000 1239.48 1239.75 1239.61 0.00 2 1000 1131.86 1132.16 1131.97 0.00 4 1000 616.73 616.93 616.84 0.01 8 1000 622.13 622.54 622.30 0.02 16 1000 101.96 102.37 102.20 0.30 32 1000 699.61 699.97 699.79 0.09 64 1000 112.47 112.72 112.62 1.08 128 1000 124.77 125.02 124.88 1.95 256 1000 147.92 148.28 148.07 3.29 512 1000 196.63 196.90 196.79 4.96 1024 1000 283.33 283.78 283.59 6.88 2048 1000 413.03 413.37 413.18 9.45 4096 1000 803.44 804.66 804.05 9.71 8192 1000 1680.42 1683.40 1681.95 9.28 16384 1000 3342.57 3350.84 3347.33 9.33 32768 1000 6539.65 6552.15 6546.60 9.54 65536 640 13359.44 13386.07 13372.03 9.34 131072 320 32503.71 32650.85 32591.06 7.66 262144 160 80541.00 81179.47 80934.90 6.16 524288 80 172252.65 173658.82 172960.10 5.76 1048576 40 343484.08 353526.08 348981.32 5.66 2097152 20 716249.35 752846.85 743033.06 5.31 4194304 10 1364635.60 1548578.80 1482620.30 5.17 #----------------------------------------------------------------------------- # Benchmarking Exchange # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 107.63 107.71 107.67 0.00 1 1000 107.12 107.21 107.16 0.04 2 1000 108.04 108.09 108.07 0.07 4 1000 110.41 111.05 110.73 0.14 8 1000 108.59 108.59 108.59 0.28 16 1000 110.43 110.51 110.47 0.55 32 1000 113.63 113.64 113.63 1.07 64 1000 129.28 129.39 129.33 1.89 128 1000 154.70 154.82 154.76 3.15 256 1000 173.84 173.85 173.85 5.62 512 1000 262.85 262.98 262.92 7.43 1024 1000 389.18 389.34 389.26 10.03 2048 1000 597.38 597.49 597.43 13.08 4096 1000 975.08 975.29 975.19 16.02 8192 1000 1739.42 1739.49 1739.45 17.97 16384 1000 3453.90 3454.13 3454.02 18.09 32768 1000 6995.91 6996.16 6996.03 17.87 65536 640 15794.01 15798.20 15796.10 15.82 131072 320 34258.91 34268.13 34263.52 14.59 262144 160 81554.42 81604.91 81579.66 12.25 524288 80 177368.42 177457.37 177412.90 11.27 1048576 40 368040.37 368218.90 368129.64 10.86 2097152 20 748867.85 749211.80 749039.83 10.68 4194304 10 1518829.20 1519613.70 1519221.45 10.53 #----------------------------------------------------------------------------- # Benchmarking Exchange # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 117.18 117.40 117.27 0.00 1 1000 114.44 114.53 114.50 0.03 2 1000 112.74 112.85 112.80 0.07 4 1000 116.36 116.47 116.41 0.13 8 1000 115.59 115.86 115.73 0.26 16 1000 118.83 118.99 118.92 0.51 32 1000 124.61 124.72 124.66 0.98 64 1000 130.16 130.24 130.19 1.87 128 1000 152.39 152.53 152.47 3.20 256 1000 178.66 178.73 178.70 5.46 512 1000 249.31 249.43 249.38 7.83 1024 1000 380.03 380.18 380.11 10.27 2048 1000 597.60 597.78 597.70 13.07 4096 1000 958.01 958.14 958.09 16.31 8192 1000 1728.46 1728.63 1728.52 18.08 16384 1000 3348.00 3348.25 3348.11 18.67 32768 1000 6502.44 6502.91 6502.63 19.22 65536 640 17388.41 17389.74 17389.00 14.38 131072 320 47003.39 47038.61 47021.15 10.63 262144 160 94796.62 95161.69 95007.22 10.51 524288 80 225407.69 226308.36 225886.25 8.84 1048576 40 438612.52 442682.55 440982.70 9.04 2097152 20 778140.80 779696.15 779069.55 10.26 4194304 10 1915766.50 1986072.70 1958198.13 8.06 #----------------------------------------------------------------------------- # Benchmarking Exchange # ( #processes = 8 ) #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 154.62 154.81 154.70 0.00 1 1000 155.63 155.83 155.73 0.02 2 1000 156.27 156.40 156.34 0.05 4 1000 155.43 155.55 155.48 0.10 8 1000 156.28 156.58 156.47 0.19 16 1000 157.19 157.27 157.23 0.39 32 1000 163.32 163.45 163.37 0.75 64 1000 174.40 174.54 174.45 1.40 128 1000 182.81 182.95 182.90 2.67 256 1000 216.85 217.04 216.97 4.50 512 1000 291.20 291.45 291.33 6.70 1024 1000 461.06 461.36 461.24 8.47 2048 1000 852.15 852.60 852.40 9.16 4096 1000 1595.66 1597.18 1596.39 9.78 8192 1000 3264.46 3267.46 3265.92 9.56 16384 1000 6636.89 6642.78 6639.77 9.41 32768 1000 16900.39 16913.92 16910.16 7.39 65536 640 36326.17 36347.86 36340.06 6.88 131072 320 96556.36 97267.58 96989.61 5.14 262144 160 194349.21 194959.09 194675.80 5.13 524288 80 347011.84 350609.02 349188.32 5.70 1048576 40 714441.83 734088.85 727102.17 5.45 2097152 20 1466246.40 1542176.90 1510138.79 5.19 4194304 10 2887263.20 3112136.20 3019078.46 5.14 #---------------------------------------------------------------- # Benchmarking Allreduce # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.06 0.07 0.06 4 1000 93.79 93.88 93.84 8 1000 96.25 96.27 96.26 16 1000 96.63 96.65 96.64 32 1000 101.38 101.45 101.41 64 1000 108.23 108.25 108.24 128 1000 118.64 118.64 118.64 256 1000 145.24 145.41 145.32 512 1000 191.44 191.53 191.48 1024 1000 282.12 282.14 282.13 2048 1000 412.71 412.93 412.82 4096 1000 589.49 589.72 589.61 8192 1000 992.34 992.45 992.39 16384 1000 1772.97 1773.04 1773.00 32768 1000 3464.36 3464.41 3464.39 65536 640 7228.57 7229.06 7228.81 131072 320 17003.92 17013.43 17008.67 262144 160 43354.48 43404.04 43379.26 524288 80 92656.61 92733.78 92695.19 1048576 40 192170.02 192319.87 192244.95 2097152 20 396210.55 396574.75 396392.65 4194304 10 794821.40 795689.90 795255.65 #---------------------------------------------------------------- # Benchmarking Allreduce # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.06 0.06 0.06 4 1000 192.83 192.94 192.88 8 1000 194.29 194.43 194.36 16 1000 194.43 194.64 194.53 32 1000 202.88 203.03 202.95 64 1000 215.38 215.59 215.49 128 1000 241.92 242.01 241.96 256 1000 278.14 278.36 278.24 512 1000 373.43 373.46 373.45 1024 1000 555.97 556.21 556.08 2048 1000 801.03 801.32 801.17 4096 1000 1155.86 1156.05 1155.97 8192 1000 1929.55 1929.66 1929.61 16384 1000 3453.35 3453.57 3453.47 32768 1000 6630.22 6630.61 6630.44 65536 640 17052.61 17055.29 17054.18 131072 320 32391.80 32402.81 32396.37 262144 160 77924.63 77980.98 77947.50 524288 80 191566.30 192071.55 191810.13 1048576 40 377506.10 377897.05 377747.26 2097152 20 768464.45 769339.95 768944.83 4194304 10 1548456.80 1552856.80 1550813.58 #---------------------------------------------------------------- # Benchmarking Allreduce # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.06 0.11 0.07 4 1000 269.70 269.83 269.76 8 1000 271.33 271.46 271.39 16 1000 273.97 274.16 274.08 32 1000 280.64 280.94 280.80 64 1000 300.23 300.41 300.32 128 1000 323.37 323.53 323.45 256 1000 377.11 377.21 377.16 512 1000 478.50 478.60 478.54 1024 1000 683.59 683.73 683.66 2048 1000 1042.59 1042.71 1042.65 4096 1000 1753.79 1753.98 1753.86 8192 1000 3408.73 3409.16 3408.90 16384 1000 7601.08 7602.87 7601.59 32768 1000 21373.07 21374.34 21373.67 65536 640 64326.81 64337.35 64331.36 131072 320 99533.68 100265.38 99892.62 262144 160 192950.60 193233.92 193070.57 524288 80 346451.01 347026.81 346730.15 1048576 40 678144.13 678884.62 678579.24 2097152 20 1439777.25 1440878.40 1440391.01 4194304 10 2852540.90 2853691.70 2853209.15 #---------------------------------------------------------------- # Benchmarking Reduce # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.09 0.10 0.09 4 1000 83.49 83.59 83.54 8 1000 84.28 84.39 84.33 16 1000 87.77 87.89 87.83 32 1000 89.47 89.57 89.52 64 1000 102.28 102.39 102.34 128 1000 119.63 119.77 119.70 256 1000 134.46 134.61 134.54 512 1000 181.29 183.40 182.34 1024 1000 274.76 275.06 274.91 2048 1000 396.82 397.21 397.02 4096 1000 575.19 575.78 575.49 8192 1000 960.75 961.73 961.24 16384 1000 1703.95 1705.65 1704.80 32768 1000 3243.32 3246.60 3244.96 65536 640 6612.43 6622.38 6617.41 131072 320 13468.62 13497.89 13483.26 262144 160 27246.06 27310.11 27278.09 524288 80 53982.76 54113.29 54048.02 1048576 40 106780.63 107148.40 106964.51 2097152 20 213451.05 214668.25 214059.65 4194304 10 426579.00 430869.60 428724.30 #---------------------------------------------------------------- # Benchmarking Reduce # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.09 0.11 0.10 4 1000 1163.52 1163.81 1163.66 8 1000 160.71 160.87 160.80 16 1000 166.43 166.78 166.61 32 1000 167.74 167.98 167.87 64 1000 184.68 185.01 184.85 128 1000 212.60 212.96 212.78 256 1000 265.17 265.62 265.40 512 1000 358.87 359.43 359.16 1024 1000 537.76 538.65 538.21 2048 1000 759.35 760.37 759.87 4096 1000 1128.83 1130.59 1129.71 8192 1000 1864.76 1867.55 1866.17 16384 1000 3318.64 3323.54 3321.09 32768 1000 6394.13 6403.39 6398.83 65536 640 12631.75 12659.60 12645.82 131072 320 25735.81 25809.59 25765.08 262144 160 51728.59 51992.40 51843.45 524288 80 101878.70 102713.16 102263.86 1048576 40 200948.05 203863.95 202286.96 2097152 20 396893.10 408070.85 402200.38 4194304 10 763785.30 814918.60 790365.67 #---------------------------------------------------------------- # Benchmarking Reduce # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.09 0.13 0.11 4 1000 1775.16 1775.61 1775.40 8 1000 1025.73 1050.07 1046.85 16 1000 623.24 623.45 623.34 32 1000 221.64 222.13 221.89 64 1000 246.64 247.01 246.82 128 1000 292.98 293.37 293.17 256 1000 367.76 368.32 368.04 512 1000 511.11 512.00 511.59 1024 1000 772.82 774.00 773.46 2048 1000 1164.32 1166.23 1165.37 4096 1000 1911.59 1914.69 1913.34 8192 1000 3408.44 3413.78 3411.51 16384 1000 6420.58 6431.12 6426.58 32768 1000 12684.95 12704.36 12696.19 65536 640 25257.51 25317.99 25292.39 131072 320 50452.19 50643.51 50560.95 262144 160 112563.42 113275.96 112963.67 524288 80 223545.21 225449.06 224558.05 1048576 40 419973.77 426636.98 423427.20 2097152 20 795087.50 833679.80 818260.53 4194304 10 1538791.50 1662602.60 1609360.95 #---------------------------------------------------------------- # Benchmarking Reduce_scatter # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 102.78 102.80 102.79 4 1000 105.34 105.47 105.41 8 1000 101.85 101.87 101.86 16 1000 105.40 105.43 105.41 32 1000 113.79 113.84 113.81 64 1000 114.99 115.04 115.01 128 1000 121.45 121.54 121.50 256 1000 133.66 133.80 133.73 512 1000 156.83 156.92 156.87 1024 1000 194.04 194.09 194.07 2048 1000 282.09 282.21 282.15 4096 1000 412.64 412.82 412.73 8192 1000 590.43 590.62 590.52 16384 1000 994.12 994.22 994.17 32768 1000 1777.42 1777.56 1777.49 65536 640 3517.40 3517.76 3517.58 131072 320 7256.79 7257.40 7257.09 262144 160 17053.72 17072.16 17062.94 524288 80 41974.77 42073.25 42024.01 1048576 40 91453.55 91613.75 91533.65 2097152 20 191756.50 192056.75 191906.63 4194304 10 395173.00 395914.40 395543.70 #---------------------------------------------------------------- # Benchmarking Reduce_scatter # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 183.95 184.15 184.03 4 1000 189.20 189.35 189.25 8 1000 189.40 189.50 189.47 16 1000 193.01 193.16 193.08 32 1000 196.69 196.81 196.76 64 1000 206.39 206.51 206.45 128 1000 226.43 226.60 226.52 256 1000 254.85 255.60 255.19 512 1000 312.89 313.01 312.94 1024 1000 406.39 406.50 406.45 2048 1000 555.75 555.96 555.86 4096 1000 817.60 817.76 817.66 8192 1000 1181.80 1182.06 1181.90 16384 1000 1742.37 1742.68 1742.53 32768 1000 2909.86 2910.10 2910.01 65536 640 6445.79 6447.34 6446.51 131072 320 14770.86 14780.29 14774.76 262144 160 34763.15 34772.09 34768.87 524288 80 73426.45 73604.64 73504.47 1048576 40 167440.42 168206.08 167775.73 2097152 20 327696.35 331566.20 329802.78 4194304 10 726034.50 735908.00 730637.35 #---------------------------------------------------------------- # Benchmarking Reduce_scatter # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 276.08 276.22 276.15 4 1000 286.23 286.37 286.28 8 1000 286.81 286.90 286.84 16 1000 288.01 288.21 288.11 32 1000 296.61 296.80 296.71 64 1000 314.50 314.78 314.65 128 1000 335.00 335.25 335.09 256 1000 377.17 377.41 377.30 512 1000 466.47 466.58 466.51 1024 1000 865.41 865.70 865.54 2048 1000 1022.66 1022.90 1022.75 4096 1000 1331.00 1331.32 1331.13 8192 1000 1842.71 1843.09 1842.90 16384 1000 2727.48 2727.79 2727.63 32768 1000 4801.47 4801.91 4801.70 65536 640 10020.45 10022.41 10021.33 131072 320 21995.53 22005.17 22001.52 262144 160 75501.79 75583.10 75545.99 524288 80 173626.49 173836.72 173757.91 1048576 40 264759.07 265279.55 264988.27 2097152 20 529908.45 532878.90 531158.53 4194304 10 1088371.20 1094599.40 1091415.06 #---------------------------------------------------------------- # Benchmarking Allgather # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.09 0.11 0.10 1 1000 94.63 94.65 94.64 2 1000 94.64 94.67 94.65 4 1000 92.27 92.35 92.31 8 1000 95.21 95.27 95.24 16 1000 96.24 96.29 96.27 32 1000 100.01 100.06 100.03 64 1000 107.87 107.94 107.90 128 1000 118.74 118.88 118.81 256 1000 143.24 143.31 143.27 512 1000 191.17 191.29 191.23 1024 1000 285.11 285.26 285.18 2048 1000 408.44 408.63 408.54 4096 1000 586.95 587.00 586.97 8192 1000 988.21 988.41 988.31 16384 1000 1758.69 1758.81 1758.75 32768 1000 3449.57 3450.04 3449.80 65536 640 7060.74 7060.93 7060.84 131072 320 16941.21 16950.71 16945.96 262144 160 41648.98 41700.38 41674.68 524288 80 90423.55 90507.46 90465.51 1048576 40 187433.05 187598.35 187515.70 2097152 20 383007.10 383298.60 383152.85 4194304 10 774329.60 774717.30 774523.45 #---------------------------------------------------------------- # Benchmarking Allgather # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.09 0.11 0.10 1 1000 178.68 178.80 178.75 2 1000 179.21 179.30 179.26 4 1000 180.70 180.84 180.76 8 1000 182.45 182.66 182.54 16 1000 186.43 186.52 186.48 32 1000 199.13 199.40 199.25 64 1000 217.07 217.22 217.15 128 1000 248.38 248.50 248.45 256 1000 317.90 318.10 318.00 512 1000 448.45 448.52 448.49 1024 1000 662.75 662.83 662.79 2048 1000 965.06 965.29 965.19 4096 1000 1533.68 1533.98 1533.77 8192 1000 2674.65 2675.02 2674.78 16384 1000 5030.06 5030.49 5030.24 32768 1000 10191.25 10191.73 10191.45 65536 640 25219.99 25224.67 25222.23 131072 320 70750.32 70798.62 70770.97 262144 160 141913.96 141996.64 141949.81 524288 80 272073.78 272255.41 272186.84 1048576 40 563153.85 563632.33 563414.33 2097152 20 1136783.95 1138430.20 1137668.34 4194304 10 2309756.60 2317867.90 2313917.58 #---------------------------------------------------------------- # Benchmarking Allgather # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.09 0.12 0.10 1 1000 267.99 268.10 268.04 2 1000 267.49 267.57 267.54 4 1000 269.74 270.12 269.94 8 1000 271.52 271.74 271.61 16 1000 279.67 279.99 279.81 32 1000 293.85 293.91 293.87 64 1000 313.41 313.63 313.52 128 1000 352.64 352.79 352.74 256 1000 443.36 443.50 443.42 512 1000 585.93 586.12 586.01 1024 1000 911.87 912.00 911.95 2048 1000 1406.21 1406.35 1406.28 4096 1000 2672.21 2672.69 2672.43 8192 1000 5679.43 5681.71 5680.52 16384 1000 11087.70 11089.69 11088.72 32768 1000 49091.76 49096.66 49093.78 65536 640 77859.93 77900.64 77880.86 131072 320 156493.50 156615.46 156555.11 262144 160 278314.98 278461.97 278384.66 524288 80 541096.56 541537.16 541277.62 1048576 40 1103292.30 1105455.05 1104345.47 2097152 20 2221152.35 2228312.50 2224679.57 4194304 10 4510391.10 4553237.80 4530901.74 #---------------------------------------------------------------- # Benchmarking Allgatherv # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.11 0.12 0.11 1 1000 99.62 99.65 99.64 2 1000 100.44 100.51 100.47 4 1000 101.93 101.98 101.96 8 1000 102.84 102.91 102.88 16 1000 107.30 107.42 107.36 32 1000 109.66 109.67 109.67 64 1000 116.40 116.40 116.40 128 1000 131.67 131.75 131.71 256 1000 148.08 148.08 148.08 512 1000 196.54 196.61 196.57 1024 1000 289.64 289.79 289.72 2048 1000 417.85 418.05 417.95 4096 1000 600.35 600.65 600.50 8192 1000 1010.30 1010.42 1010.36 16384 1000 1812.78 1812.86 1812.82 32768 1000 3768.37 3768.45 3768.41 65536 640 7878.30 7880.59 7879.45 131072 320 18163.38 18172.10 18167.74 262144 160 41775.89 41827.27 41801.58 524288 80 90520.51 90607.45 90563.98 1048576 40 187932.42 188105.45 188018.94 2097152 20 383052.60 383394.60 383223.60 4194304 10 771240.50 772004.20 771622.35 #---------------------------------------------------------------- # Benchmarking Allgatherv # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.11 0.13 0.11 1 1000 185.51 185.54 185.53 2 1000 186.40 186.49 186.43 4 1000 187.25 187.27 187.26 8 1000 189.55 189.72 189.64 16 1000 194.40 194.63 194.51 32 1000 206.55 206.64 206.60 64 1000 227.25 227.38 227.32 128 1000 260.54 260.67 260.59 256 1000 329.64 329.76 329.70 512 1000 458.81 458.94 458.86 1024 1000 673.69 673.96 673.78 2048 1000 979.38 979.72 979.50 4096 1000 1557.83 1558.12 1557.93 8192 1000 2735.77 2736.01 2735.93 16384 1000 5310.57 5310.97 5310.73 32768 1000 11070.77 11071.34 11070.97 65536 640 26702.23 26707.98 26705.32 131072 320 71887.22 71921.99 71903.00 262144 160 146053.48 146190.78 146122.48 524288 80 285141.43 285780.25 285433.35 1048576 40 571423.65 573813.03 572541.71 2097152 20 1160742.50 1170176.80 1165374.65 4194304 10 2320998.20 2359145.70 2340134.97 #---------------------------------------------------------------- # Benchmarking Allgatherv # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.11 0.19 0.13 1 1000 277.14 277.28 277.22 2 1000 277.80 277.83 277.82 4 1000 276.76 277.00 276.86 8 1000 281.37 281.44 281.39 16 1000 288.12 288.22 288.17 32 1000 300.77 300.91 300.84 64 1000 323.91 323.99 323.93 128 1000 369.65 369.75 369.71 256 1000 468.81 469.11 468.96 512 1000 628.32 628.73 628.50 1024 1000 933.21 933.58 933.37 2048 1000 1435.37 1435.51 1435.43 4096 1000 2740.25 2740.51 2740.33 8192 1000 5874.95 5877.88 5876.35 16384 1000 11554.00 11555.08 11554.63 32768 1000 40874.34 40875.31 40874.86 65536 640 93545.97 93579.02 93565.27 131072 320 227115.16 227205.33 227164.77 262144 160 560890.82 561413.11 561158.52 524288 80 1192590.49 1194856.81 1193936.87 1048576 40 2401698.27 2407103.23 2404264.09 2097152 20 4888262.50 4923894.25 4907784.71 4194304 10 9923179.00 10079250.40 10027351.75 #---------------------------------------------------------------- # Benchmarking Alltoall # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 98.59 98.72 98.65 1 1000 99.93 100.05 99.99 2 1000 103.03 103.14 103.09 4 1000 102.60 102.72 102.66 8 1000 104.75 104.80 104.78 16 1000 110.54 110.62 110.58 32 1000 115.45 115.51 115.48 64 1000 123.85 123.87 123.86 128 1000 121.22 121.30 121.26 256 1000 145.18 145.33 145.25 512 1000 193.38 193.45 193.41 1024 1000 281.85 281.93 281.89 2048 1000 406.95 407.10 407.03 4096 1000 585.86 586.12 585.99 8192 1000 981.70 981.79 981.75 16384 1000 1756.63 1756.74 1756.68 32768 1000 3470.11 3470.14 3470.12 65536 640 7033.07 7033.11 7033.09 131072 320 17213.51 17222.73 17218.12 262144 160 41781.84 41834.61 41808.22 524288 80 90866.74 90953.85 90910.29 1048576 40 188052.17 188226.05 188139.11 2097152 20 383494.15 383847.85 383671.00 4194304 10 775287.60 776074.20 775680.90 #---------------------------------------------------------------- # Benchmarking Alltoall # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 194.28 194.38 194.33 1 1000 199.00 199.11 199.05 2 1000 199.43 199.51 199.48 4 1000 201.43 201.60 201.52 8 1000 214.49 214.72 214.59 16 1000 237.56 237.67 237.61 32 1000 263.31 263.47 263.41 64 1000 151.05 151.09 151.07 128 1000 173.23 173.33 173.28 256 1000 217.35 217.42 217.38 512 1000 307.75 307.86 307.83 1024 1000 483.38 483.46 483.41 2048 1000 803.85 803.93 803.89 4096 1000 1345.24 1345.38 1345.30 8192 1000 2515.65 2515.80 2515.72 16384 1000 5561.74 5564.63 5563.22 32768 1000 16253.06 16260.12 16257.52 65536 640 27455.91 27458.28 27456.95 131072 320 73010.38 73053.76 73038.12 262144 160 170401.09 170593.82 170486.19 524288 80 338083.10 339075.04 338626.29 1048576 40 725270.87 727787.45 726453.26 2097152 20 1458399.10 1467883.45 1462993.10 4194304 10 2966855.40 3005952.50 2986681.85 #---------------------------------------------------------------- # Benchmarking Alltoall # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 271.99 272.06 272.02 1 1000 281.31 281.39 281.35 2 1000 285.53 285.68 285.60 4 1000 298.59 298.77 298.67 8 1000 319.84 319.95 319.88 16 1000 365.90 365.98 365.95 32 1000 373.71 373.86 373.79 64 1000 385.48 385.62 385.57 128 1000 411.62 411.80 411.70 256 1000 502.43 502.66 502.54 512 1000 730.33 730.51 730.44 1024 1000 1218.40 1218.83 1218.65 2048 1000 2600.39 2601.69 2601.09 4096 1000 4830.37 4832.76 4831.64 8192 1000 9774.02 9779.59 9777.14 16384 1000 26881.95 26890.57 26886.92 32768 1000 68142.90 68149.62 68147.70 65536 640 166248.59 166279.13 166263.24 131072 320 286205.92 286305.88 286253.69 262144 160 512190.06 512534.28 512363.00 524288 80 1053482.00 1054664.56 1053955.35 1048576 40 2169504.05 2171445.75 2170699.37 2097152 20 4279971.60 4295622.75 4289078.31 4194304 10 8713639.80 8727401.60 8722372.61 #---------------------------------------------------------------- # Benchmarking Bcast # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.07 0.07 0.07 1 1000 81.41 81.45 81.43 2 1000 82.74 82.78 82.76 4 1000 83.46 83.54 83.50 8 1000 83.25 83.33 83.29 16 1000 84.46 84.52 84.49 32 1000 89.43 89.50 89.47 64 1000 95.39 95.45 95.42 128 1000 118.21 118.33 118.27 256 1000 132.63 132.75 132.69 512 1000 188.38 188.54 188.46 1024 1000 275.18 275.41 275.30 2048 1000 395.90 396.25 396.07 4096 1000 571.56 572.06 571.81 8192 1000 951.65 952.53 952.09 16384 1000 1765.63 1766.15 1765.89 32768 1000 3340.00 3340.44 3340.22 65536 640 6460.25 6461.14 6460.70 131072 320 12674.00 12676.50 12675.25 262144 160 36739.38 36793.52 36766.45 524288 80 73110.66 73212.47 73161.57 1048576 40 145167.72 145325.60 145246.66 2097152 20 288447.95 288770.05 288609.00 4194304 10 578733.60 579380.30 579056.95 #---------------------------------------------------------------- # Benchmarking Bcast # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.07 0.07 0.07 1 1000 1148.24 1148.41 1148.30 2 1000 226.45 226.63 226.50 4 1000 96.86 97.09 96.94 8 1000 97.22 97.37 97.26 16 1000 98.41 98.57 98.46 32 1000 103.26 103.45 103.33 64 1000 115.88 116.06 115.94 128 1000 133.25 133.48 133.34 256 1000 169.70 169.93 169.79 512 1000 240.06 240.41 240.20 1024 1000 372.12 372.74 372.41 2048 1000 568.53 569.50 569.02 4096 1000 931.63 932.81 932.26 8192 1000 1646.17 1648.14 1647.28 16384 1000 2844.77 2845.05 2844.85 32768 1000 5089.81 5090.10 5089.95 65536 640 9697.74 9698.04 9697.90 131072 320 33837.88 33859.82 33849.23 262144 160 49040.98 49110.46 49075.77 524288 80 95416.96 95616.39 95507.26 1048576 40 199134.17 199818.45 199525.94 2097152 20 410270.70 411892.35 411199.75 4194304 10 832762.60 834591.10 833771.90 #---------------------------------------------------------------- # Benchmarking Bcast # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.07 0.09 0.07 1 1000 1566.60 1566.87 1566.70 2 1000 1028.95 1029.54 1029.29 4 1000 734.74 763.65 749.20 8 1000 495.69 496.21 495.97 16 1000 125.99 126.25 126.15 32 1000 130.33 130.54 130.42 64 1000 142.74 142.99 142.85 128 1000 160.33 160.62 160.45 256 1000 189.17 189.55 189.35 512 1000 259.56 260.04 259.84 1024 1000 392.16 393.28 392.72 2048 1000 592.13 593.43 592.83 4096 1000 970.49 972.48 971.59 8192 1000 1718.94 1722.36 1720.82 16384 1000 2698.56 2698.75 2698.65 32768 1000 5157.23 5158.67 5157.88 65536 640 10226.60 10229.64 10227.70 131072 320 35422.18 35437.27 35430.29 262144 160 82933.37 83045.29 82995.53 524288 80 122856.84 123002.78 122929.26 1048576 40 197367.65 197887.73 197558.35 2097152 20 387076.85 389241.95 387949.13 4194304 10 782607.30 797596.80 789578.49 #--------------------------------------------------- # Benchmarking Barrier # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #--------------------------------------------------- #repetitions t_min[usec] t_max[usec] t_avg[usec] 1000 90.66 90.71 90.69 #--------------------------------------------------- # Benchmarking Barrier # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #--------------------------------------------------- #repetitions t_min[usec] t_max[usec] t_avg[usec] 1000 188.27 188.38 188.32 #--------------------------------------------------- # Benchmarking Barrier # ( #processes = 8 ) #--------------------------------------------------- #repetitions t_min[usec] t_max[usec] t_avg[usec] 1000 262.52 262.77 262.65 #===================================================== # # Thanks for using PMB2.2 # # The Pallas team kindly requests that you # give us as much feedback for PMB as possible. # # It would be very helpful when you sent the # output tables of your run(s) of PMB to # # ####################### # # # # # pmb@pallas.com # # # # # ####################### # # You might also add # # - personal information (institution, motivation # for using PMB) # - basic information about the machine you used # (number of CPUs, processor type e.t.c.) # #=====================================================