#--------------------------------------------------- # PALLAS MPI Benchmark Suite V2.2, MPI-1 part #--------------------------------------------------- # Date : Wed Mar 24 19:42:32 2004 # Machine : i686# System : Linux # Release : 2.4.20-8smp # Version : #1 SMP Thu Mar 13 16:43:01 EST 2003 # # 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 164.39 0.00 1 1000 47.58 0.02 2 1000 49.61 0.04 4 1000 50.76 0.08 8 1000 49.88 0.15 16 1000 49.53 0.31 32 1000 186.38 0.16 64 1000 52.11 1.17 128 1000 87.57 1.39 256 1000 70.82 3.45 512 1000 63.90 7.64 1024 1000 76.55 12.76 2048 1000 100.68 19.40 4096 1000 35436.52 0.11 8192 1000 19599.41 0.40 16384 1000 243.05 64.29 32768 1000 4622.75 6.76 65536 640 849.85 73.54 131072 320 3792.39 32.96 262144 160 3770.59 66.30 524288 80 7379.44 67.76 1048576 40 14511.75 68.91 2097152 20 31348.88 63.80 4194304 10 62547.30 63.95 #--------------------------------------------------- # Benchmarking PingPing # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #--------------------------------------------------- #bytes #repetitions t[usec] Mbytes/sec 0 1000 56.00 0.00 1 1000 56.03 0.02 2 1000 55.46 0.03 4 1000 54.00 0.07 8 1000 55.55 0.14 16 1000 56.52 0.27 32 1000 55.91 0.55 64 1000 57.64 1.06 128 1000 59.93 2.04 256 1000 66.66 3.66 512 1000 72.32 6.75 1024 1000 88.35 11.05 2048 1000 112.90 17.30 4096 1000 147.48 26.49 8192 1000 213.80 36.54 16384 1000 343.83 45.44 32768 1000 703.80 44.40 65536 640 2022.78 30.90 131072 320 3349.89 37.31 262144 160 6436.38 38.84 524288 80 12828.23 38.98 1048576 40 25260.80 39.59 2097152 20 55249.35 36.20 4194304 10 113319.70 35.30 #----------------------------------------------------------------------------- # 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 57.13 57.21 57.17 0.00 1 1000 56.57 56.58 56.57 0.03 2 1000 54.90 54.91 54.91 0.07 4 1000 55.29 55.30 55.30 0.14 8 1000 55.66 55.71 55.69 0.27 16 1000 55.75 55.77 55.76 0.55 32 1000 56.27 56.35 56.31 1.08 64 1000 57.12 57.21 57.16 2.13 128 1000 59.15 59.18 59.17 4.13 256 1000 65.18 65.21 65.20 7.49 512 1000 73.06 73.11 73.09 13.36 1024 1000 89.28 89.32 89.30 21.87 2048 1000 112.14 112.18 112.16 34.82 4096 1000 145.86 145.86 145.86 53.56 8192 1000 211.92 211.98 211.95 73.71 16384 1000 343.01 343.21 343.11 91.05 32768 1000 957.40 957.58 957.49 65.27 65536 640 2301.44 2301.60 2301.52 54.31 131072 320 3523.35 3523.49 3523.42 70.95 262144 160 6412.78 6414.26 6413.52 77.95 524288 80 12779.41 12785.83 12782.62 78.21 1048576 40 25309.15 25329.17 25319.16 78.96 2097152 20 54971.95 55009.50 54990.73 72.71 4194304 10 110029.20 110168.70 110098.95 72.62 #----------------------------------------------------------------------------- # 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 1940.68 1940.76 1940.73 0.00 1 1000 1703.98 1704.03 1704.00 0.00 2 1000 1599.50 1599.55 1599.51 0.00 4 1000 1706.56 1706.63 1706.60 0.00 8 1000 1706.26 1706.36 1706.29 0.01 16 1000 1586.81 1625.55 1596.50 0.02 32 1000 968.24 968.29 968.27 0.06 64 1000 816.52 816.54 816.53 0.15 128 1000 56.20 56.31 56.27 4.34 256 1000 133.06 133.14 133.10 3.67 512 1000 73.08 73.17 73.13 13.35 1024 1000 91.63 91.75 91.67 21.29 2048 1000 121.42 121.55 121.48 32.14 4096 1000 156.20 156.30 156.23 49.98 8192 1000 208.76 208.92 208.83 74.79 16384 1000 347.07 347.18 347.15 90.01 32768 1000 1240.86 1241.09 1240.94 50.36 65536 640 1764.16 1765.79 1765.21 70.79 131072 320 4137.90 4144.54 4141.71 60.32 262144 160 7705.86 7722.76 7714.29 64.74 524288 80 15334.35 15408.95 15372.84 64.90 1048576 40 25879.77 26165.10 26023.01 76.44 2097152 20 57062.45 58340.85 57710.25 68.56 4194304 10 116141.50 121123.30 118649.75 66.05 #----------------------------------------------------------------------------- # Benchmarking Sendrecv # ( #processes = 8 ) #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 2150.65 2189.72 2179.65 0.00 1 1000 2140.76 2141.13 2140.93 0.00 2 1000 2024.60 2062.51 2056.89 0.00 4 1000 1830.49 1830.77 1830.66 0.00 8 1000 1599.54 1599.90 1599.71 0.01 16 1000 1739.69 1739.86 1739.76 0.02 32 1000 1318.07 1318.29 1318.17 0.05 64 1000 1399.23 1399.43 1399.33 0.09 128 1000 325.32 325.53 325.42 0.75 256 1000 71.62 71.72 71.66 6.81 512 1000 83.40 83.63 83.54 11.68 1024 1000 101.84 102.02 101.90 19.14 2048 1000 133.89 134.21 134.03 29.10 4096 1000 182.34 182.60 182.47 42.79 8192 1000 288.46 289.12 288.84 54.04 16384 1000 9841.16 9841.74 9841.46 3.18 32768 1000 10546.30 10547.54 10546.93 5.93 65536 640 5333.57 5340.96 5338.13 23.40 131072 320 5955.53 5974.57 5966.70 41.84 262144 160 11742.23 11828.48 11790.70 42.27 524288 80 22849.17 23179.20 23028.46 43.14 1048576 40 43779.88 44844.60 44459.63 44.60 2097152 20 95581.45 101585.50 98843.52 39.38 4194304 10 171173.30 190785.30 181251.44 41.93 #----------------------------------------------------------------------------- # 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 74.70 74.72 74.71 0.00 1 1000 74.60 74.60 74.60 0.05 2 1000 75.21 75.24 75.22 0.10 4 1000 76.58 76.60 76.59 0.20 8 1000 75.21 75.24 75.22 0.41 16 1000 76.09 76.11 76.10 0.80 32 1000 76.00 76.04 76.02 1.61 64 1000 77.64 77.70 77.67 3.14 128 1000 79.34 79.34 79.34 6.15 256 1000 84.37 84.39 84.38 11.57 512 1000 97.18 97.26 97.22 20.08 1024 1000 118.62 118.62 118.62 32.93 2048 1000 160.49 160.55 160.52 48.66 4096 1000 287.32 287.41 287.37 54.36 8192 1000 338.68 338.71 338.70 92.26 16384 1000 667.77 667.79 667.78 93.59 32768 1000 1514.75 1514.97 1514.86 82.51 65536 640 3435.06 3435.28 3435.17 72.77 131072 320 6509.26 6509.50 6509.38 76.81 262144 160 12943.24 12944.76 12944.00 77.25 524288 80 25975.31 25981.53 25978.42 76.98 1048576 40 50579.15 50597.08 50588.11 79.06 2097152 20 111227.15 111283.00 111255.08 71.89 4194304 10 221873.60 222032.90 221953.25 72.06 #----------------------------------------------------------------------------- # 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 77.95 78.06 78.01 0.00 1 1000 77.74 77.81 77.78 0.05 2 1000 77.37 77.44 77.41 0.10 4 1000 78.87 78.96 78.92 0.19 8 1000 78.09 78.14 78.12 0.39 16 1000 78.68 78.72 78.70 0.78 32 1000 78.64 78.71 78.67 1.55 64 1000 79.61 79.65 79.63 3.07 128 1000 83.08 83.13 83.11 5.87 256 1000 87.78 87.87 87.82 11.11 512 1000 104.14 104.21 104.17 18.74 1024 1000 125.08 125.18 125.13 31.21 2048 1000 444.18 444.26 444.23 17.59 4096 1000 532.72 532.79 532.76 29.33 8192 1000 449.63 449.80 449.69 69.48 16384 1000 1029.58 1029.72 1029.68 60.70 32768 1000 2224.01 2225.18 2224.68 56.18 65536 640 4571.08 4573.28 4572.42 54.67 131072 320 9763.25 9775.24 9770.41 51.15 262144 160 16783.44 16831.97 16811.62 59.41 524288 80 32464.39 32642.31 32569.00 61.27 1048576 40 64329.10 65053.60 64752.20 61.49 2097152 20 140734.60 143867.35 142599.68 55.61 4194304 10 277739.30 290268.80 285282.40 55.12 #----------------------------------------------------------------------------- # Benchmarking Exchange # ( #processes = 8 ) #----------------------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] Mbytes/sec 0 1000 121.00 121.20 121.10 0.00 1 1000 99.52 99.62 99.56 0.04 2 1000 100.73 100.91 100.83 0.08 4 1000 99.92 100.00 99.95 0.15 8 1000 99.59 99.77 99.67 0.31 16 1000 100.61 100.76 100.67 0.61 32 1000 100.05 100.16 100.11 1.22 64 1000 102.74 102.87 102.82 2.37 128 1000 106.02 106.22 106.13 4.60 256 1000 113.64 113.73 113.70 8.59 512 1000 131.04 131.27 131.16 14.88 1024 1000 155.51 155.77 155.65 25.08 2048 1000 223.82 224.01 223.91 34.88 4096 1000 383.13 383.40 383.28 40.75 8192 1000 2555.77 2556.23 2556.01 12.23 16384 1000 11326.62 11366.59 11360.96 5.50 32768 1000 20681.21 20682.53 20681.85 6.04 65536 640 8120.15 8127.79 8124.77 30.76 131072 320 15601.49 15632.12 15620.01 31.99 262144 160 24051.04 24132.29 24088.19 41.44 524288 80 46669.05 47002.82 46895.69 42.55 1048576 40 93778.88 96313.25 95362.71 41.53 2097152 20 197516.00 203764.25 200738.30 39.26 4194304 10 387033.70 413176.20 403196.94 38.72 #---------------------------------------------------------------- # 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 61.53 61.53 61.53 8 1000 61.81 61.81 61.81 16 1000 61.64 61.65 61.64 32 1000 61.55 61.55 61.55 64 1000 65.71 65.77 65.74 128 1000 66.26 66.28 66.27 256 1000 71.77 71.83 71.80 512 1000 81.56 81.61 81.58 1024 1000 99.76 99.83 99.80 2048 1000 123.08 123.10 123.09 4096 1000 155.05 155.07 155.06 8192 1000 231.03 231.10 231.07 16384 1000 401.94 402.04 401.99 32768 1000 894.46 894.56 894.51 65536 640 2099.97 2100.20 2100.09 131072 320 4532.81 4533.09 4532.95 262144 160 8825.35 8827.21 8826.28 524288 80 16978.68 16981.86 16980.27 1048576 40 34183.75 34207.60 34195.68 2097152 20 73904.30 73967.20 73935.75 4194304 10 145663.90 145700.40 145682.15 #---------------------------------------------------------------- # 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 113.16 113.22 113.20 8 1000 114.31 114.44 114.38 16 1000 113.85 113.91 113.88 32 1000 114.73 114.77 114.75 64 1000 116.48 116.53 116.51 128 1000 121.49 121.59 121.54 256 1000 130.44 130.49 130.46 512 1000 159.11 159.34 159.22 1024 1000 198.80 199.01 198.90 2048 1000 237.36 237.40 237.38 4096 1000 331.67 331.98 331.83 8192 1000 467.01 467.26 467.14 16384 1000 781.36 781.48 781.43 32768 1000 1658.82 1658.98 1658.91 65536 640 3830.97 3831.50 3831.27 131072 320 8197.95 8198.48 8198.17 262144 160 16155.58 16157.43 16156.78 524288 80 31614.05 31628.16 31620.31 1048576 40 62844.00 62864.70 62851.42 2097152 20 136022.95 136192.10 136100.59 4194304 10 267803.60 268084.20 267941.33 #---------------------------------------------------------------- # Benchmarking Allreduce # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.06 0.06 0.06 4 1000 210.12 210.27 210.20 8 1000 190.49 190.60 190.54 16 1000 194.92 195.04 194.97 32 1000 197.06 197.19 197.11 64 1000 205.04 205.14 205.09 128 1000 242.68 242.90 242.75 256 1000 211.88 212.00 211.93 512 1000 240.78 240.94 240.83 1024 1000 297.00 297.22 297.08 2048 1000 361.49 361.65 361.56 4096 1000 509.19 509.46 509.31 8192 1000 2397.62 2397.78 2397.70 16384 1000 8533.66 8533.99 8533.86 32768 1000 5377.55 5377.99 5377.78 65536 640 10480.98 10481.86 10481.47 131072 320 33255.76 33259.11 33257.51 262144 160 39372.88 39388.48 39383.26 524288 80 67297.55 67343.95 67329.10 1048576 40 132972.40 133188.20 133099.36 2097152 20 262398.15 262962.75 262775.60 4194304 10 529030.20 531370.80 530673.34 #---------------------------------------------------------------- # 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 0.12 0.10 4 1000 49.14 49.17 49.15 8 1000 50.12 50.14 50.13 16 1000 326.87 326.90 326.89 32 1000 49.57 49.63 49.60 64 1000 52.26 52.30 52.28 128 1000 53.40 53.44 53.42 256 1000 90.19 90.25 90.22 512 1000 63.22 63.27 63.25 1024 1000 87.47 87.56 87.51 2048 1000 104.53 104.63 104.58 4096 1000 33843.78 33845.03 33844.41 8192 1000 16000.55 16001.19 16000.87 16384 1000 272.03 272.20 272.11 32768 1000 5970.23 5970.64 5970.43 65536 640 1485.87 1486.60 1486.23 131072 320 4148.68 4152.03 4150.36 262144 160 6390.51 6404.48 6397.49 524288 80 12601.11 12656.94 12629.02 1048576 40 24200.70 24415.73 24308.21 2097152 20 50756.60 51632.70 51194.65 4194304 10 98258.30 101584.90 99921.60 #---------------------------------------------------------------- # 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 0.09 0.09 4 1000 1225.65 1225.83 1225.76 8 1000 1076.52 1076.61 1076.57 16 1000 1046.79 1046.92 1046.86 32 1000 362.57 362.66 362.63 64 1000 97.95 98.10 98.04 128 1000 104.46 104.63 104.54 256 1000 112.76 112.87 112.82 512 1000 129.52 129.68 129.61 1024 1000 167.31 167.60 167.46 2048 1000 205.92 206.19 206.06 4096 1000 80281.93 80284.77 80283.18 8192 1000 24331.81 24333.06 24332.48 16384 1000 543.36 543.98 543.66 32768 1000 12360.92 12598.56 12479.77 65536 640 2490.83 2493.40 2492.06 131072 320 6609.96 6622.82 6616.36 262144 160 11176.21 11231.59 11203.80 524288 80 21522.41 21739.71 21630.91 1048576 40 42670.02 43503.58 43082.33 2097152 20 86975.35 90501.85 88736.98 4194304 10 164781.10 178608.80 171712.43 #---------------------------------------------------------------- # Benchmarking Reduce # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.09 0.13 0.10 4 1000 2041.50 2041.90 2041.75 8 1000 1608.59 1608.93 1608.80 16 1000 1274.57 1274.88 1274.74 32 1000 1258.28 1258.57 1258.43 64 1000 2521.69 2521.96 2521.86 128 1000 1154.05 1154.41 1154.25 256 1000 237.95 238.24 238.12 512 1000 475.32 475.77 475.55 1024 1000 565.46 565.86 565.66 2048 1000 1054.84 1055.52 1055.22 4096 1000 11095.48 11096.18 11095.88 8192 1000 6244.91 6245.89 6245.47 16384 1000 1016.69 1018.04 1017.34 32768 1000 4701.10 4703.30 4702.22 65536 640 5229.53 5238.41 5234.71 131072 320 9990.81 10032.07 10014.04 262144 160 22359.65 22520.77 22443.39 524288 80 39522.14 40138.21 39839.96 1048576 40 79146.33 81682.65 80470.95 2097152 20 156791.95 167023.45 162128.12 4194304 10 297489.00 336709.10 317704.37 #---------------------------------------------------------------- # 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 66.55 66.56 66.56 4 1000 68.40 68.49 68.44 8 1000 71.48 71.50 71.49 16 1000 72.12 72.17 72.14 32 1000 70.18 70.21 70.19 64 1000 71.33 71.39 71.36 128 1000 73.78 73.81 73.79 256 1000 73.97 73.97 73.97 512 1000 80.14 80.16 80.15 1024 1000 80.52 80.57 80.54 2048 1000 96.61 96.61 96.61 4096 1000 124.87 124.94 124.91 8192 1000 154.93 154.99 154.96 16384 1000 233.98 234.06 234.02 32768 1000 414.66 414.75 414.71 65536 640 951.88 952.10 951.99 131072 320 2464.98 2465.23 2465.11 262144 160 4253.32 4253.62 4253.47 524288 80 8839.79 8843.54 8841.66 1048576 40 17426.10 17439.77 17432.94 2097152 20 34332.80 34367.80 34350.30 4194304 10 75868.20 75976.70 75922.45 #---------------------------------------------------------------- # 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 119.58 119.61 119.60 4 1000 123.95 123.99 123.97 8 1000 124.98 125.09 125.03 16 1000 125.07 125.13 125.11 32 1000 126.79 126.87 126.82 64 1000 126.70 126.73 126.71 128 1000 133.21 133.33 133.27 256 1000 137.42 137.50 137.46 512 1000 148.57 148.71 148.64 1024 1000 196.12 196.14 196.13 2048 1000 227.42 227.52 227.46 4096 1000 283.83 284.01 283.92 8192 1000 358.15 358.21 358.17 16384 1000 875.76 875.86 875.81 32768 1000 1003.17 1003.44 1003.28 65536 640 1403.26 1403.44 1403.33 131072 320 5143.20 5143.54 5143.34 262144 160 7902.69 7909.34 7906.40 524288 80 13710.16 13734.51 13723.40 1048576 40 28218.55 28289.05 28254.06 2097152 20 55943.20 55975.90 55956.21 4194304 10 108377.90 108426.40 108404.70 #---------------------------------------------------------------- # Benchmarking Reduce_scatter # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 198.04 198.17 198.10 4 1000 215.85 215.95 215.90 8 1000 228.13 228.28 228.17 16 1000 235.41 235.46 235.44 32 1000 223.80 223.90 223.85 64 1000 214.31 214.34 214.33 128 1000 214.88 215.08 214.97 256 1000 264.17 264.44 264.25 512 1000 249.20 249.26 249.22 1024 1000 522.36 522.51 522.41 2048 1000 561.81 561.89 561.84 4096 1000 631.80 631.91 631.85 8192 1000 729.02 729.21 729.08 16384 1000 953.32 953.46 953.40 32768 1000 1654.73 1654.89 1654.80 65536 640 8137.91 8138.99 8138.52 131072 320 54293.38 54296.82 54294.77 262144 160 105510.46 105516.00 105513.95 524288 80 47515.19 47554.15 47535.98 1048576 40 56595.92 56618.63 56608.55 2097152 20 99279.85 99538.55 99412.62 4194304 10 172352.70 172652.90 172508.99 #---------------------------------------------------------------- # Benchmarking Allgather # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.10 0.11 0.11 1 1000 61.08 61.12 61.10 2 1000 61.36 61.41 61.39 4 1000 60.94 60.97 60.96 8 1000 62.12 62.13 62.12 16 1000 60.44 60.45 60.44 32 1000 62.14 62.17 62.15 64 1000 62.91 62.92 62.92 128 1000 64.45 64.50 64.47 256 1000 69.80 69.86 69.83 512 1000 82.45 82.49 82.47 1024 1000 97.86 97.91 97.89 2048 1000 120.60 120.68 120.64 4096 1000 152.53 152.54 152.53 8192 1000 226.12 226.20 226.16 16384 1000 389.10 389.14 389.12 32768 1000 959.02 959.18 959.10 65536 640 2246.65 2246.77 2246.71 131072 320 3790.00 3790.38 3790.19 262144 160 7605.22 7607.68 7606.45 524288 80 14541.09 14545.74 14543.41 1048576 40 28782.13 28795.70 28788.91 2097152 20 63013.20 63050.45 63031.82 4194304 10 128475.30 128623.90 128549.60 #---------------------------------------------------------------- # 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 0.10 0.10 1 1000 113.88 113.95 113.92 2 1000 114.09 114.24 114.17 4 1000 112.98 113.14 113.05 8 1000 114.67 114.73 114.69 16 1000 115.45 115.49 115.47 32 1000 116.79 116.90 116.84 64 1000 119.88 119.91 119.90 128 1000 126.71 126.78 126.74 256 1000 143.74 143.79 143.76 512 1000 178.94 179.20 179.06 1024 1000 215.75 215.94 215.84 2048 1000 287.75 287.86 287.81 4096 1000 399.90 400.02 399.96 8192 1000 632.94 633.13 633.03 16384 1000 1205.41 1205.52 1205.46 32768 1000 2578.03 2578.17 2578.08 65536 640 5390.43 5390.95 5390.68 131072 320 10175.50 10176.37 10175.98 262144 160 20111.88 20115.16 20113.57 524288 80 40087.57 40096.93 40090.93 1048576 40 83948.25 83980.97 83960.04 2097152 20 173802.05 173955.90 173862.35 4194304 10 340482.80 341904.50 341429.10 #---------------------------------------------------------------- # Benchmarking Allgather # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.09 0.10 0.10 1 1000 189.54 189.67 189.61 2 1000 188.54 188.67 188.58 4 1000 189.71 189.78 189.75 8 1000 196.16 196.28 196.23 16 1000 214.11 214.19 214.14 32 1000 202.14 202.20 202.17 64 1000 225.60 225.80 225.72 128 1000 221.78 221.89 221.82 256 1000 228.64 228.71 228.68 512 1000 272.46 272.68 272.53 1024 1000 365.93 366.05 365.99 2048 1000 508.27 508.48 508.37 4096 1000 1456.16 1456.35 1456.25 8192 1000 3262.55 3263.05 3262.75 16384 1000 6181.64 6182.35 6181.94 32768 1000 13278.26 13278.91 13278.63 65536 640 20265.56 20267.43 20266.82 131072 320 28088.45 28095.04 28092.45 262144 160 57892.69 57915.79 57906.71 524288 80 110447.49 110485.70 110467.45 1048576 40 217008.65 217341.40 217226.60 2097152 20 431995.00 432933.75 432665.51 4194304 10 852077.30 855157.80 854048.23 #---------------------------------------------------------------- # 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 64.89 64.93 64.91 2 1000 65.87 65.93 65.90 4 1000 65.27 65.30 65.28 8 1000 64.87 64.87 64.87 16 1000 66.67 66.67 66.67 32 1000 65.67 65.70 65.69 64 1000 67.44 67.47 67.46 128 1000 70.74 70.75 70.74 256 1000 75.39 75.44 75.42 512 1000 85.81 85.85 85.83 1024 1000 102.96 103.01 102.99 2048 1000 130.85 130.90 130.88 4096 1000 165.66 165.69 165.68 8192 1000 251.87 251.92 251.90 16384 1000 469.39 469.62 469.50 32768 1000 1404.15 1404.32 1404.23 65536 640 3168.91 3169.07 3168.99 131072 320 5167.69 5167.91 5167.80 262144 160 7525.44 7527.46 7526.45 524288 80 14493.23 14496.21 14494.72 1048576 40 28954.13 28965.40 28959.76 2097152 20 64090.50 64149.75 64120.13 4194304 10 127566.50 127781.60 127674.05 #---------------------------------------------------------------- # 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 0.12 0.11 1 1000 117.90 117.96 117.93 2 1000 119.64 119.74 119.68 4 1000 119.09 119.15 119.12 8 1000 118.03 118.10 118.06 16 1000 121.00 121.12 121.07 32 1000 121.34 121.42 121.38 64 1000 126.20 126.31 126.25 128 1000 130.11 130.19 130.16 256 1000 152.37 152.45 152.41 512 1000 185.70 185.91 185.80 1024 1000 227.50 227.62 227.57 2048 1000 298.65 298.74 298.70 4096 1000 426.40 426.48 426.44 8192 1000 706.49 706.80 706.61 16384 1000 1657.25 1657.42 1657.34 32768 1000 3656.20 3656.46 3656.29 65536 640 7409.27 7410.09 7409.72 131072 320 12709.97 12710.71 12710.33 262144 160 24732.26 24765.27 24752.44 524288 80 44277.45 44289.45 44285.19 1048576 40 81637.87 81920.65 81781.22 2097152 20 181650.60 182952.35 182301.24 4194304 10 370772.00 376060.20 373430.13 #---------------------------------------------------------------- # Benchmarking Allgatherv # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 0.11 0.13 0.12 1 1000 219.54 219.70 219.61 2 1000 208.40 208.47 208.42 4 1000 228.33 228.41 228.36 8 1000 197.43 197.51 197.48 16 1000 200.08 200.19 200.12 32 1000 244.76 244.84 244.79 64 1000 208.16 208.30 208.23 128 1000 218.39 218.43 218.40 256 1000 274.98 275.09 275.04 512 1000 310.03 310.30 310.12 1024 1000 373.25 373.48 373.36 2048 1000 541.53 541.75 541.60 4096 1000 1567.70 1568.06 1567.85 8192 1000 3766.66 3799.13 3782.90 16384 1000 8120.31 8120.86 8120.64 32768 1000 14339.56 14371.06 14355.39 65536 640 43799.90 43865.75 43826.31 131072 320 45893.69 45914.07 45905.22 262144 160 87448.49 87499.22 87478.72 524288 80 165157.11 165393.84 165318.01 1048576 40 334936.92 336721.12 336127.11 2097152 20 674434.55 678268.25 676973.39 4194304 10 1355213.40 1365962.00 1362015.90 #---------------------------------------------------------------- # Benchmarking Alltoall # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 65.22 65.24 65.23 1 1000 66.49 66.56 66.52 2 1000 65.62 65.62 65.62 4 1000 66.82 66.89 66.86 8 1000 66.63 66.64 66.64 16 1000 65.32 65.36 65.34 32 1000 67.32 67.37 67.35 64 1000 69.96 69.99 69.97 128 1000 63.62 63.65 63.64 256 1000 68.48 68.54 68.51 512 1000 77.73 77.78 77.76 1024 1000 96.78 96.83 96.81 2048 1000 118.16 118.22 118.19 4096 1000 150.26 150.28 150.27 8192 1000 219.33 219.35 219.34 16384 1000 388.16 388.17 388.16 32768 1000 975.07 975.08 975.07 65536 640 2338.84 2339.01 2338.93 131072 320 4777.49 4777.78 4777.64 262144 160 7576.37 7579.20 7577.79 524288 80 14513.08 14517.98 14515.53 1048576 40 34756.22 34768.15 34762.19 2097152 20 63427.00 63471.80 63449.40 4194304 10 126419.20 127827.10 127123.15 #---------------------------------------------------------------- # Benchmarking Alltoall # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 117.91 117.96 117.93 1 1000 118.77 118.83 118.80 2 1000 120.22 120.25 120.23 4 1000 120.56 120.60 120.58 8 1000 121.88 122.04 121.95 16 1000 127.02 127.05 127.03 32 1000 132.08 132.15 132.12 64 1000 92.79 92.89 92.84 128 1000 96.47 96.55 96.50 256 1000 100.83 100.93 100.86 512 1000 115.36 115.40 115.38 1024 1000 139.34 139.47 139.39 2048 1000 196.58 196.62 196.61 4096 1000 275.60 275.71 275.65 8192 1000 1795.10 1795.19 1795.14 16384 1000 1318.93 1319.27 1319.13 32768 1000 4244.98 4246.20 4245.75 65536 640 7318.15 7318.94 7318.52 131072 320 12680.08 12686.50 12683.99 262144 160 28466.77 28486.55 28476.82 524288 80 48790.33 48861.10 48825.28 1048576 40 94283.10 94658.85 94473.02 2097152 20 210602.45 211824.85 211221.39 4194304 10 421863.90 422111.00 422017.08 #---------------------------------------------------------------- # Benchmarking Alltoall # ( #processes = 8 ) #---------------------------------------------------------------- #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] 0 1000 207.54 207.63 207.58 1 1000 205.23 205.34 205.28 2 1000 222.10 222.23 222.15 4 1000 204.08 204.21 204.13 8 1000 208.17 208.30 208.25 16 1000 218.47 218.50 218.48 32 1000 286.89 287.04 286.97 64 1000 294.48 294.66 294.56 128 1000 292.77 292.87 292.83 256 1000 298.20 298.27 298.24 512 1000 330.29 330.42 330.35 1024 1000 359.20 359.34 359.27 2048 1000 654.40 654.54 654.47 4096 1000 1038.29 1038.57 1038.44 8192 1000 4747.93 4748.49 4748.23 16384 1000 17855.57 17857.98 17857.14 32768 1000 101009.48 101049.16 101019.93 65536 640 44944.16 44948.29 44946.21 131072 320 49552.10 49573.62 49564.66 262144 160 90348.32 90408.69 90377.36 524288 80 159108.01 159178.61 159150.20 1048576 40 312139.88 312327.20 312245.73 2097152 20 647225.75 649613.80 648626.57 4194304 10 1726678.90 1826564.50 1742065.44 #---------------------------------------------------------------- # 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 48.53 48.55 48.54 2 1000 48.30 48.36 48.33 4 1000 48.32 48.38 48.35 8 1000 48.17 48.24 48.20 16 1000 48.86 48.94 48.90 32 1000 49.29 49.36 49.33 64 1000 89.95 90.01 89.98 128 1000 89.50 89.56 89.53 256 1000 91.21 91.29 91.25 512 1000 64.85 64.94 64.90 1024 1000 83.14 83.18 83.16 2048 1000 101.19 101.26 101.22 4096 1000 42053.34 42054.57 42053.95 8192 1000 16704.83 16705.23 16705.03 16384 1000 316.21 316.21 316.21 32768 1000 576.92 576.98 576.95 65536 640 4117.14 4117.17 4117.15 131072 320 2711.54 2711.78 2711.66 262144 160 5311.24 5311.89 5311.56 524288 80 10600.41 10604.52 10602.47 1048576 40 20478.58 20489.35 20483.96 2097152 20 40249.00 40277.50 40263.25 4194304 10 88874.50 88991.00 88932.75 #---------------------------------------------------------------- # 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 796.70 796.82 796.75 2 1000 1233.02 1233.13 1233.06 4 1000 1067.11 1067.25 1067.20 8 1000 1090.47 1090.63 1090.53 16 1000 1062.33 1062.46 1062.39 32 1000 70.32 70.49 70.39 64 1000 56.70 56.88 56.77 128 1000 96.13 96.30 96.20 256 1000 64.08 64.23 64.15 512 1000 75.77 75.95 75.86 1024 1000 135.37 135.57 135.45 2048 1000 128.76 128.97 128.86 4096 1000 1970.43 1970.70 1970.58 8192 1000 518.67 519.06 518.87 16384 1000 2380.71 2380.85 2380.78 32768 1000 1065.61 1065.77 1065.68 65536 640 6522.64 6523.04 6522.90 131072 320 5817.80 5819.01 5818.29 262144 160 8273.64 8274.20 8273.87 524288 80 16224.15 16229.14 16226.98 1048576 40 30529.90 30546.22 30537.65 2097152 20 59768.65 59804.10 59788.86 4194304 10 129222.50 129341.50 129277.28 #---------------------------------------------------------------- # 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 1389.68 1389.80 1389.74 2 1000 1302.30 1302.53 1302.44 4 1000 1384.57 1384.78 1384.70 8 1000 1334.45 1334.63 1334.56 16 1000 1231.37 1265.12 1247.53 32 1000 758.34 758.48 758.41 64 1000 398.45 398.55 398.49 128 1000 86.53 86.62 86.57 256 1000 124.72 124.88 124.79 512 1000 109.70 109.81 109.75 1024 1000 130.61 130.74 130.66 2048 1000 170.42 170.66 170.52 4096 1000 243.40 244.28 244.00 8192 1000 373.59 406.54 398.15 16384 1000 4672.89 4673.06 4672.99 32768 1000 3203.36 3203.53 3203.44 65536 640 3266.37 3266.89 3266.56 131072 320 8825.37 8826.42 8825.86 262144 160 11417.28 11421.36 11419.56 524288 80 23522.05 23534.08 23527.69 1048576 40 39710.20 39752.85 39735.43 2097152 20 77189.35 77438.00 77323.57 4194304 10 150873.20 151162.20 151032.49 #--------------------------------------------------- # Benchmarking Barrier # ( #processes = 2 ) # ( 6 additional processes waiting in MPI_Barrier) #--------------------------------------------------- #repetitions t_min[usec] t_max[usec] t_avg[usec] 1000 57.94 58.00 57.97 #--------------------------------------------------- # Benchmarking Barrier # ( #processes = 4 ) # ( 4 additional processes waiting in MPI_Barrier) #--------------------------------------------------- #repetitions t_min[usec] t_max[usec] t_avg[usec] 1000 110.32 110.35 110.33 #--------------------------------------------------- # Benchmarking Barrier # ( #processes = 8 ) #--------------------------------------------------- #repetitions t_min[usec] t_max[usec] t_avg[usec] 1000 184.72 184.83 184.78 #===================================================== # # 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.) # #=====================================================