b_eff = 7129.367 MB/s = 648.124 * 11 PEs with 256 MB/PE on SUPER-UX hwwsx4 9.1 Rev1 SX-4 SECTION-HEAD-BEGIN b_eff.c, Revision 3.2 from Nov. 07, 1999 MEMORY_PER_PROCESSOR = 256 MBytes [1M = 1024*1024] 1-dim-paterns: size = 11 1-dim-paterns: size = 11 2-dim-paterns: size = 5 * 2 3-dim-paterns: size = 2 * 2 * 2 Used message lengths (and used loop length for each message lengths): msglng= 1 (300), 2 (300), 4 (300), 8 (300), 16 (300), 32 (300), 64 (300), 128 (300), 256 (300), 512 (300), 1024 (300), 2048 (300), 4096 (300), 8933 (234), 19484 (107), 42495 (49), 92682 (22), 202141 (10), 440872 (4), 961548 (2), 2097152 (1), SECTION-HEAD-END SECTION-LOOP-BEGIN measurment loop: elapsed time = 85.871 sec sum of max elapsed time per entries above = 85.180 sec difference = 0.691 sec = 0.8% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-5*2&+1 => 1 sendrecv_calls with 11 messages, i.e. 4.2 msgs/used node, all nodes are used p01 ring-3*4&-1 => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p02 ring-1*11fix => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p03 ring-1*11fix => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p04 ring-1*11fix => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p05 ring-1*11fix => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 10 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p18 worst bi-section => 2 sendrecv_calls with 10 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p19 acyclic-1dim-all => 2 sendrecv_calls with 20 messages, i.e. 4.2 msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 26 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p21 acyclic-3dim-all => 6 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, 3 nodes are UNUSED p22 cyclic-1dim-all => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p23 cyclic-2dim-all => 3 sendrecv_calls with 30 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p24 cyclic-3dim-all => 3 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, 3 nodes are UNUSED SECTION-PATTERN-END Printing accumulated bandwidth / number of processes = bandwidth per process (in MBytes/sec = 1e6 Bytes/sec) Additionally the accumulated value processes is printed in the last column Only the cyclic and random patterns are used to compute b_eff (except p00), the other patterns are only as information. SECTION-BY-METHODS-BEGIN 2nd analysis path, only as information, last row: for all mthd: logavg_pat ( avg_L ( max_rep ( b(L) ) ) ) || logavg_pat ( max_mthd ( avg_L ( max_rep ( b(L) ) ) ) ) pattern / methods: Sndrcv Alltoal non-blk -> best method -> accumulated p00 ring-5*2&+1 : 648.452 173.726 527.101 -> 648.452 -> 7132.971 MByte/s p01 ring-3*4&-1 : 653.000 315.166 534.508 -> 653.000 -> 7182.996 MByte/s p02 ring-1*11fix : 642.230 483.086 540.304 -> 642.230 -> 7064.533 MByte/s p03 ring-1*11fix : 631.072 483.631 552.921 -> 631.072 -> 6941.787 MByte/s p04 ring-1*11fix : 651.799 482.991 546.892 -> 651.799 -> 7169.793 MByte/s p05 ring-1*11fix : 650.102 483.757 554.917 -> 650.102 -> 7151.125 MByte/s p06 random-cyc-1dim : 652.927 480.394 539.550 -> 652.927 -> 7182.193 MByte/s p07 random-cyc-1dim : 646.247 482.942 529.557 -> 646.247 -> 7108.713 MByte/s p08 random-cyc-1dim : 645.687 481.302 528.403 -> 645.687 -> 7102.555 MByte/s p09 random-cyc-1dim : 649.879 482.619 529.683 -> 649.879 -> 7148.673 MByte/s p10 random-cyc-1dim : 648.470 480.801 551.689 -> 648.470 -> 7133.173 MByte/s p11 random-cyc-1dim : 650.385 484.694 544.363 -> 650.385 -> 7154.239 MByte/s p12 random-cyc-1dim : 651.208 484.295 528.300 -> 651.208 -> 7163.285 MByte/s p13 random-cyc-1dim : 635.289 482.638 555.321 -> 635.289 -> 6988.179 MByte/s p14 random-cyc-1dim : 649.350 483.861 552.658 -> 649.350 -> 7142.847 MByte/s p15 random-cyc-1dim : 651.933 482.679 544.143 -> 651.933 -> 7171.264 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 636.466 482.719 563.875 -> 636.466 -> 7001.131 MByte/s p17 best bi-section : 368.590 270.713 519.801 -> 519.801 -> 5717.813 MByte/s p18 worst bi-section : 369.402 201.333 521.634 -> 521.634 -> 5737.976 MByte/s p19 acyclic-1dim-all : 550.101 442.214 519.290 -> 550.101 -> 6051.115 MByte/s p20 acyclic-2dim-all : 417.190 377.400 447.646 -> 447.646 -> 4924.109 MByte/s p21 acyclic-3dim-all : 296.475 320.400 365.588 -> 365.588 -> 4021.465 MByte/s p22 cyclic-1dim-all : 651.530 482.082 541.163 -> 651.530 -> 7166.834 MByte/s p23 cyclic-2dim-all : 579.350 432.699 485.240 -> 579.350 -> 6372.848 MByte/s p24 cyclic-3dim-all : 477.726 322.730 402.428 -> 477.726 -> 5254.988 MByte/s log_avg of all rings : 646.065 379.535 542.683 || 646.065 -> 7106.711 MByte/s log_avg of all random : 648.119 482.621 540.269 || 648.119 -> 7129.314 MByte/s log_avg(ring,random) : 647.091 427.985 541.475 ||(647.091 -> 7118.003)MByte/s SECTION-BY-METHODS-END SECTION-BY-REPETITIONS-BEGIN 3rd analysis path, only as information, last row: for all rep.:logavg_pat ( avg_L ( max_mthd ( b(L) ) ) ) || logavg_pat ( max_rep ( avg_L ( max_mthd ( b(L) ) ) ) ) pattern / repetition: rep.0 rep.1 rep.2 -> best repetition -> accumulated p00 ring-5*2&+1 : 619.958 637.220 642.282 -> 642.282 -> 7065.100 MByte/s p01 ring-3*4&-1 : 593.606 555.804 640.384 -> 640.384 -> 7044.225 MByte/s p02 ring-1*11fix : 599.294 571.446 639.544 -> 639.544 -> 7034.979 MByte/s p03 ring-1*11fix : 606.993 608.523 639.375 -> 639.375 -> 7033.125 MByte/s p04 ring-1*11fix : 603.682 604.225 624.037 -> 624.037 -> 6864.407 MByte/s p05 ring-1*11fix : 605.588 594.256 630.361 -> 630.361 -> 6933.970 MByte/s p06 random-cyc-1dim : 592.129 537.449 651.699 -> 651.699 -> 7168.690 MByte/s p07 random-cyc-1dim : 609.414 602.127 644.527 -> 644.527 -> 7089.802 MByte/s p08 random-cyc-1dim : 566.588 629.980 637.697 -> 637.697 -> 7014.663 MByte/s p09 random-cyc-1dim : 591.707 611.978 619.508 -> 619.508 -> 6814.591 MByte/s p10 random-cyc-1dim : 564.477 612.206 641.072 -> 641.072 -> 7051.790 MByte/s p11 random-cyc-1dim : 614.431 615.373 642.391 -> 642.391 -> 7066.300 MByte/s p12 random-cyc-1dim : 608.840 595.272 650.096 -> 650.096 -> 7151.056 MByte/s p13 random-cyc-1dim : 598.357 618.437 637.498 -> 637.498 -> 7012.474 MByte/s p14 random-cyc-1dim : 610.144 574.083 638.591 -> 638.591 -> 7024.501 MByte/s p15 random-cyc-1dim : 585.384 553.575 649.932 -> 649.932 -> 7149.255 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 574.216 562.523 638.979 -> 638.979 -> 7028.771 MByte/s p17 best bi-section : 502.442 520.509 521.540 -> 521.540 -> 5736.935 MByte/s p18 worst bi-section : 517.218 475.610 517.036 -> 517.218 -> 5689.397 MByte/s p19 acyclic-1dim-all : 553.538 528.258 569.962 -> 569.962 -> 6269.586 MByte/s p20 acyclic-2dim-all : 447.983 435.103 454.754 -> 454.754 -> 5002.293 MByte/s p21 acyclic-3dim-all : 354.964 340.422 349.818 -> 354.964 -> 3904.609 MByte/s p22 cyclic-1dim-all : 557.614 558.607 649.165 -> 649.165 -> 7140.810 MByte/s p23 cyclic-2dim-all : 538.013 529.349 561.779 -> 561.779 -> 6179.569 MByte/s p24 cyclic-3dim-all : 440.703 434.791 477.125 -> 477.125 -> 5248.373 MByte/s log_avg of all rings : 604.800 594.665 635.963 || 635.963 -> 6995.594 MByte/s log_avg of all random : 593.904 594.331 641.240 || 641.240 -> 7053.635 MByte/s log_avg(ring,random) : 599.327 594.498 638.596 ||(638.596 -> 7024.555)MByte/s SECTION-BY-REPETITIONS-END SECTION-BY-MTHD-MSGLNG-BEGIN 4th analysis path, only as information, last 3 rows: for all methods and some msglng L: logavg_pat ( max_rep ( b(L) ) ) for all methods: logavg_pat ( avg_L ( max_rep ( b(L) ) ) ) pattern & method / msg: 1 16 256 4096 92682 2097152 -> average -> accumulated p00 ring-5*2&+1 p00 method 0 : 0.051 0.845 13.365 164.692 1451.798 3511.876 -> 648.452 -> 7132.971 MByte/s p00 method 1 : 0.007 0.132 2.089 33.901 407.857 862.648 -> 173.726 -> 1910.985 MByte/s p00 method 2 : 0.035 0.559 8.987 127.103 1215.105 2646.262 -> 527.101 -> 5798.112 MByte/s p01 ring-3*4&-1 p01 method 0 : 0.051 0.852 13.426 165.361 1457.785 3568.236 -> 653.000 -> 7182.996 MByte/s p01 method 1 : 0.014 0.262 4.172 66.167 781.268 1402.462 -> 315.166 -> 3466.827 MByte/s p01 method 2 : 0.040 0.671 10.441 138.966 1369.018 2644.620 -> 534.508 -> 5879.590 MByte/s p02 ring-1*11fix p02 method 0 : 0.050 0.851 13.413 158.557 1443.510 3562.852 -> 642.230 -> 7064.533 MByte/s p02 method 1 : 0.014 0.264 4.140 67.101 793.534 3154.883 -> 483.086 -> 5313.941 MByte/s p02 method 2 : 0.039 0.670 10.465 142.131 1368.286 2640.731 -> 540.304 -> 5943.345 MByte/s p03 ring-1*11fix p03 method 0 : 0.051 0.851 13.426 158.500 1452.601 3564.111 -> 631.072 -> 6941.787 MByte/s p03 method 1 : 0.014 0.265 4.187 67.133 790.314 3152.872 -> 483.631 -> 5319.943 MByte/s p03 method 2 : 0.040 0.668 10.449 137.110 1368.676 2609.592 -> 552.921 -> 6082.127 MByte/s p04 ring-1*11fix p04 method 0 : 0.052 0.851 13.410 165.308 1458.436 3581.202 -> 651.799 -> 7169.793 MByte/s p04 method 1 : 0.013 0.266 4.191 65.301 792.414 3138.979 -> 482.991 -> 5312.903 MByte/s p04 method 2 : 0.040 0.659 10.486 139.167 1369.783 2655.523 -> 546.892 -> 6015.807 MByte/s p05 ring-1*11fix p05 method 0 : 0.051 0.850 13.421 160.814 1440.903 3575.853 -> 650.102 -> 7151.125 MByte/s p05 method 1 : 0.014 0.265 4.193 67.049 791.244 3134.138 -> 483.757 -> 5321.322 MByte/s p05 method 2 : 0.039 0.667 10.450 142.050 1370.390 2841.515 -> 554.917 -> 6104.092 MByte/s p06 random-cyc-1dim p06 method 0 : 0.051 0.851 13.417 165.121 1459.546 3563.120 -> 652.927 -> 7182.193 MByte/s p06 method 1 : 0.013 0.265 4.199 64.296 787.815 3144.590 -> 480.394 -> 5284.333 MByte/s p06 method 2 : 0.039 0.663 10.491 141.958 1365.947 2593.790 -> 539.550 -> 5935.048 MByte/s p07 random-cyc-1dim p07 method 0 : 0.049 0.818 12.910 154.795 1453.682 3518.287 -> 646.247 -> 7108.713 MByte/s p07 method 1 : 0.014 0.263 4.194 66.444 790.841 3131.087 -> 482.942 -> 5312.366 MByte/s p07 method 2 : 0.039 0.664 10.340 137.165 1365.786 2412.250 -> 529.557 -> 5825.126 MByte/s p08 random-cyc-1dim p08 method 0 : 0.048 0.818 12.895 159.859 1450.010 3531.201 -> 645.687 -> 7102.555 MByte/s p08 method 1 : 0.014 0.260 4.199 66.731 782.057 3111.779 -> 481.302 -> 5294.323 MByte/s p08 method 2 : 0.039 0.662 10.341 136.849 1370.184 2435.637 -> 528.403 -> 5812.434 MByte/s p09 random-cyc-1dim p09 method 0 : 0.051 0.850 13.417 164.738 1435.235 3533.748 -> 649.879 -> 7148.673 MByte/s p09 method 1 : 0.013 0.266 4.194 66.213 788.468 3133.820 -> 482.619 -> 5308.811 MByte/s p09 method 2 : 0.039 0.660 10.481 136.206 1368.643 2681.026 -> 529.683 -> 5826.508 MByte/s p10 random-cyc-1dim p10 method 0 : 0.051 0.851 13.416 160.001 1435.215 3544.211 -> 648.470 -> 7133.173 MByte/s p10 method 1 : 0.014 0.264 4.204 66.416 786.016 3097.146 -> 480.801 -> 5288.809 MByte/s p10 method 2 : 0.040 0.660 10.479 135.005 1365.534 2823.200 -> 551.689 -> 6068.579 MByte/s p11 random-cyc-1dim p11 method 0 : 0.051 0.851 13.421 161.682 1457.756 3541.434 -> 650.385 -> 7154.239 MByte/s p11 method 1 : 0.013 0.266 4.193 66.504 789.698 3147.421 -> 484.694 -> 5331.637 MByte/s p11 method 2 : 0.039 0.659 10.438 130.024 1361.180 2662.009 -> 544.363 -> 5987.994 MByte/s p12 random-cyc-1dim p12 method 0 : 0.051 0.852 13.414 158.011 1446.020 3561.836 -> 651.208 -> 7163.285 MByte/s p12 method 1 : 0.013 0.266 4.197 66.437 792.064 3140.935 -> 484.295 -> 5327.246 MByte/s p12 method 2 : 0.040 0.669 10.495 141.810 1370.133 2407.067 -> 528.300 -> 5811.305 MByte/s p13 random-cyc-1dim p13 method 0 : 0.050 0.821 12.942 152.781 1442.219 3536.345 -> 635.289 -> 6988.179 MByte/s p13 method 1 : 0.013 0.266 4.196 65.182 788.333 3133.633 -> 482.638 -> 5309.013 MByte/s p13 method 2 : 0.039 0.670 10.450 137.302 1369.374 2871.631 -> 555.321 -> 6108.532 MByte/s p14 random-cyc-1dim p14 method 0 : 0.050 0.821 12.964 159.791 1442.183 3570.715 -> 649.350 -> 7142.847 MByte/s p14 method 1 : 0.014 0.264 4.204 62.637 788.638 3148.158 -> 483.861 -> 5322.467 MByte/s p14 method 2 : 0.039 0.670 10.458 142.037 1367.237 2769.614 -> 552.658 -> 6079.242 MByte/s p15 random-cyc-1dim p15 method 0 : 0.050 0.852 13.430 160.140 1461.409 3538.637 -> 651.933 -> 7171.264 MByte/s p15 method 1 : 0.013 0.265 4.199 66.052 788.516 3134.587 -> 482.679 -> 5309.471 MByte/s p15 method 2 : 0.040 0.668 10.440 138.243 1362.781 2834.096 -> 544.143 -> 5985.573 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.051 0.851 13.421 165.148 1449.400 3570.228 -> 636.466 -> 7001.131 MByte/s p16 method 1 : 0.014 0.265 4.179 66.098 783.613 3148.858 -> 482.719 -> 5309.904 MByte/s p16 method 2 : 0.039 0.663 10.343 137.218 1360.246 3152.758 -> 563.875 -> 6202.623 MByte/s p17 best bi-section p17 method 0 : 0.035 0.590 9.220 106.231 763.048 2219.009 -> 368.590 -> 4054.487 MByte/s p17 method 1 : 0.006 0.122 1.926 30.615 441.482 1430.028 -> 270.713 -> 2977.846 MByte/s p17 method 2 : 0.025 0.402 6.297 90.081 989.216 3164.802 -> 519.801 -> 5717.813 MByte/s p18 worst bi-section p18 method 0 : 0.035 0.590 9.201 106.124 763.818 2219.340 -> 369.402 -> 4063.426 MByte/s p18 method 1 : 0.006 0.121 1.929 30.188 444.705 541.649 -> 201.333 -> 2214.663 MByte/s p18 method 2 : 0.025 0.404 6.337 88.889 990.082 3144.341 -> 521.634 -> 5737.976 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.046 0.775 12.213 150.280 1325.547 2618.087 -> 550.101 -> 6051.115 MByte/s p19 method 1 : 0.012 0.241 3.832 60.259 728.970 2854.728 -> 442.214 -> 4864.354 MByte/s p19 method 2 : 0.033 0.586 9.495 126.901 1244.007 2580.496 -> 519.290 -> 5712.186 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.034 0.566 8.902 106.049 911.957 2428.741 -> 417.190 -> 4589.093 MByte/s p20 method 1 : 0.016 0.321 5.072 80.302 813.568 2039.160 -> 377.400 -> 4151.405 MByte/s p20 method 2 : 0.028 0.455 6.792 98.824 1048.433 2273.718 -> 447.646 -> 4924.109 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.029 0.482 7.514 85.843 602.216 1779.704 -> 296.475 -> 3261.228 MByte/s p21 method 1 : 0.016 0.346 5.399 85.338 791.298 1550.640 -> 320.400 -> 3524.405 MByte/s p21 method 2 : 0.019 0.334 5.014 74.073 851.222 1804.090 -> 365.588 -> 4021.465 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.051 0.852 13.416 164.565 1446.127 3548.217 -> 651.530 -> 7166.834 MByte/s p22 method 1 : 0.014 0.264 4.162 66.990 786.744 3119.017 -> 482.082 -> 5302.900 MByte/s p22 method 2 : 0.040 0.668 10.486 137.245 1363.831 2649.913 -> 541.163 -> 5952.795 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.046 0.776 12.224 149.717 1326.340 3231.286 -> 579.350 -> 6372.848 MByte/s p23 method 1 : 0.018 0.372 5.848 93.149 924.333 2345.882 -> 432.699 -> 4759.684 MByte/s p23 method 2 : 0.035 0.552 9.316 124.714 1319.164 2353.745 -> 485.240 -> 5337.641 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.037 0.620 9.594 120.305 1065.535 2585.111 -> 477.726 -> 5254.988 MByte/s p24 method 1 : 0.016 0.350 5.502 87.297 810.012 1547.188 -> 322.730 -> 3550.026 MByte/s p24 method 2 : 0.026 0.435 7.073 97.327 975.432 2101.855 -> 402.428 -> 4426.707 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.850 13.410 162.177 1450.824 3560.615 || 646.065 -> 7106.711 MByte/s - ring, method 1 : 0.012 0.236 3.721 59.471 707.385 2215.917 || 379.535 -> 4174.881 MByte/s - ring, method 2 : 0.039 0.648 10.197 137.657 1342.248 2671.974 || 542.683 -> 5969.517 MByte/s log_avg of all random - random, method 0 : 0.050 0.838 13.220 159.650 1448.298 3543.919 || 648.119 -> 7129.314 MByte/s - random, method 1 : 0.013 0.264 4.198 65.679 788.240 3132.278 || 482.621 -> 5308.826 MByte/s - random, method 2 : 0.040 0.664 10.441 137.615 1366.677 2643.427 || 540.269 -> 5942.963 MByte/s log_avg(ring,random) - average, method 0 : 0.051 0.844 13.315 160.909 1449.560 3552.257 || 647.091 -> 7118.003 MByte/s - average, method 1 : 0.013 0.250 3.952 62.498 746.719 2634.552 || 427.985 -> 4707.836 MByte/s - average, method 2 : 0.039 0.656 10.318 137.636 1354.407 2657.662 || 541.475 -> 5956.225 MByte/s SECTION-BY-MTHD-MSGLNG-END SECTION-BY-MSGLNG-BEGIN 5th analysis path only as information: logavg_pat ( max_mthd ( max_rep ( b(L) ) ) ) and for all methods: logavg_pat ( max_rep ( b(L) ) ) msg length|accumulated| effective bandwidth per process: msg length| effective| average rings random methd_1 methd_2 methd_3 | bandwidth| crt,rnd only only Sendrcv Alltoall non-blk 1 0.557 0.051 0.051 0.050 0.051 0.013 0.039 2 1.101 0.100 0.101 0.099 0.100 0.025 0.077 4 2.263 0.206 0.208 0.204 0.206 0.061 0.161 8 4.631 0.421 0.425 0.417 0.421 0.122 0.327 16 9.286 0.844 0.850 0.838 0.844 0.250 0.656 32 18.498 1.682 1.695 1.668 1.682 0.495 1.311 64 36.921 3.356 3.380 3.333 3.356 0.998 2.621 128 73.184 6.653 6.709 6.598 6.653 1.975 5.180 256 146.464 13.315 13.410 13.220 13.315 3.952 10.318 512 292.766 26.615 26.821 26.411 26.615 7.927 20.591 1024 582.598 52.963 53.370 52.560 52.963 15.839 41.167 2048 899.489 81.772 82.330 81.217 81.772 31.684 69.279 4096 1769.994 160.909 162.177 159.650 160.909 62.498 137.636 8933 2329.417 211.765 212.188 211.343 211.765 110.739 187.579 19484 6084.087 553.099 557.560 548.674 553.099 221.703 513.394 42495 9131.205 830.110 833.247 826.984 830.110 400.786 762.044 92682 15945.165 1449.560 1450.824 1448.298 1449.560 746.719 1354.407 202141 18521.019 1683.729 1682.112 1685.348 1683.729 1233.406 1514.543 440872 30978.615 2816.238 2812.299 2820.182 2816.238 1759.495 2379.402 961548 23796.542 2163.322 2141.313 2185.557 2139.027 1696.118 1705.844 2097152 39074.831 3552.257 3560.615 3543.919 3552.257 2634.552 2657.662 SECTION-BY-MSGLNG-END SECTION-BY-PATTERN-MSGLNG-BEGIN official (1st) analysis path -- see last column: (only columns of some message lengths are printed) logavg_pat ( avg_L ( max_mthd ( max_rep ( b(L) ) ) ) ) pattern / msg-length: 1 16 256 4096 92682 2097152 -> average -> accumulated p00 ring-5*2&+1 : 0.051 0.845 13.365 164.692 1451.798 3511.876 -> 648.452 -> 7132.971 MByte/s p01 ring-3*4&-1 : 0.051 0.852 13.426 165.361 1457.785 3568.236 -> 653.000 -> 7182.996 MByte/s p02 ring-1*11fix : 0.050 0.851 13.413 158.557 1443.510 3562.852 -> 642.230 -> 7064.533 MByte/s p03 ring-1*11fix : 0.051 0.851 13.426 158.500 1452.601 3564.111 -> 641.235 -> 7053.580 MByte/s p04 ring-1*11fix : 0.052 0.851 13.410 165.308 1458.436 3581.202 -> 651.799 -> 7169.793 MByte/s p05 ring-1*11fix : 0.051 0.850 13.421 160.814 1440.903 3575.853 -> 650.102 -> 7151.125 MByte/s p06 random-cyc-1dim : 0.051 0.851 13.417 165.121 1459.546 3563.120 -> 652.927 -> 7182.193 MByte/s p07 random-cyc-1dim : 0.049 0.818 12.910 154.795 1453.682 3518.287 -> 646.247 -> 7108.713 MByte/s p08 random-cyc-1dim : 0.048 0.818 12.895 159.859 1450.010 3531.201 -> 645.687 -> 7102.555 MByte/s p09 random-cyc-1dim : 0.051 0.850 13.417 164.738 1435.235 3533.748 -> 649.879 -> 7148.673 MByte/s p10 random-cyc-1dim : 0.051 0.851 13.416 160.001 1435.215 3544.211 -> 648.470 -> 7133.173 MByte/s p11 random-cyc-1dim : 0.051 0.851 13.421 161.682 1457.756 3541.434 -> 650.385 -> 7154.239 MByte/s p12 random-cyc-1dim : 0.051 0.852 13.414 158.011 1446.020 3561.836 -> 651.208 -> 7163.285 MByte/s p13 random-cyc-1dim : 0.050 0.821 12.942 152.781 1442.219 3536.345 -> 638.650 -> 7025.146 MByte/s p14 random-cyc-1dim : 0.050 0.821 12.964 159.791 1442.183 3570.715 -> 649.350 -> 7142.847 MByte/s p15 random-cyc-1dim : 0.050 0.852 13.430 160.140 1461.409 3538.637 -> 651.933 -> 7171.264 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.051 0.851 13.421 165.148 1449.400 3570.228 -> 644.073 -> 7084.808 MByte/s p17 best bi-section : 0.035 0.590 9.220 106.231 989.216 3164.802 -> 522.067 -> 5742.740 MByte/s p18 worst bi-section : 0.035 0.590 9.201 106.124 990.082 3144.341 -> 523.936 -> 5763.294 MByte/s p19 acyclic-1dim-all : 0.046 0.775 12.213 150.280 1325.547 2854.728 -> 572.922 -> 6302.145 MByte/s p20 acyclic-2dim-all : 0.034 0.566 8.902 106.049 1048.433 2428.741 -> 456.729 -> 5024.022 MByte/s p21 acyclic-3dim-all : 0.029 0.482 7.514 85.843 851.222 1804.090 -> 368.302 -> 4051.325 MByte/s p22 cyclic-1dim-all : 0.051 0.852 13.416 164.565 1446.127 3548.217 -> 651.530 -> 7166.834 MByte/s p23 cyclic-2dim-all : 0.046 0.776 12.224 149.717 1326.340 3231.286 -> 579.350 -> 6372.848 MByte/s p24 cyclic-3dim-all : 0.037 0.620 9.594 120.305 1065.535 2585.111 -> 477.726 -> 5254.988 MByte/s log_avg of all rings : 0.051 0.850 13.410 162.177 1450.824 3560.615 || 647.787 -> 7125.659 MByte/s log_avg of all random : 0.050 0.838 13.220 159.650 1448.298 3543.919 || 648.462 -> 7133.077 MByte/s log_avg(ring,random) : 0.051 0.844 13.315 160.909 1449.560 3552.257 || 648.124 -> 7129.367 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 7129.367 MByte/s on 11 processes ( = 648.124 MByte/s * 11 processes) system parameters : 11 nodes, 256 MB/node system name: SUPER-UX hostname : hwwsx4 OS release : 9.1 OS version : Rev1 machine : SX-4 SECTION-BEFF-END b_eff = 7129.367 MB/s = 648.124 * 11 PEs with 256 MB/PE on SUPER-UX hwwsx4 9.1 Rev1 SX-4