b_eff = 5765.670 MB/s = 640.630 * 9 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 = 9 1-dim-paterns: size = 9 2-dim-paterns: size = 3 * 3 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 = 79.503 sec sum of max elapsed time per entries above = 78.719 sec difference = 0.784 sec = 1.0% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-4*2&+1 => 1 sendrecv_calls with 9 messages, i.e. 4.2 msgs/used node, all nodes are used p01 ring-2*4&+1 => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p02 ring-1*9fix => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p03 ring-1*9fix => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p04 ring-1*9fix => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p05 ring-1*9fix => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 8 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p18 worst bi-section => 2 sendrecv_calls with 8 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p19 acyclic-1dim-all => 2 sendrecv_calls with 16 messages, i.e. 4.2 msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p22 cyclic-1dim-all => 2 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 36 messages, i.e. 4.2 msgs/used node, all nodes are used p24 cyclic-3dim-all => 3 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, 1 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-4*2&+1 : 648.224 178.427 526.151 -> 648.224 -> 5834.018 MByte/s p01 ring-2*4&+1 : 617.135 325.034 522.537 -> 617.135 -> 5554.213 MByte/s p02 ring-1*9fix : 647.602 497.482 544.032 -> 647.602 -> 5828.419 MByte/s p03 ring-1*9fix : 649.015 499.821 559.231 -> 649.015 -> 5841.131 MByte/s p04 ring-1*9fix : 616.015 499.748 539.043 -> 616.015 -> 5544.134 MByte/s p05 ring-1*9fix : 646.521 499.333 564.055 -> 646.521 -> 5818.687 MByte/s p06 random-cyc-1dim : 631.226 499.761 539.920 -> 631.226 -> 5681.035 MByte/s p07 random-cyc-1dim : 635.338 496.847 557.611 -> 635.338 -> 5718.043 MByte/s p08 random-cyc-1dim : 627.502 499.442 520.268 -> 627.502 -> 5647.521 MByte/s p09 random-cyc-1dim : 645.864 498.845 538.408 -> 645.864 -> 5812.775 MByte/s p10 random-cyc-1dim : 646.788 499.041 565.664 -> 646.788 -> 5821.093 MByte/s p11 random-cyc-1dim : 629.655 500.639 549.748 -> 629.655 -> 5666.892 MByte/s p12 random-cyc-1dim : 634.030 499.457 557.472 -> 634.030 -> 5706.270 MByte/s p13 random-cyc-1dim : 647.046 500.713 528.461 -> 647.046 -> 5823.411 MByte/s p14 random-cyc-1dim : 636.867 498.465 554.067 -> 636.867 -> 5731.805 MByte/s p15 random-cyc-1dim : 607.312 502.239 543.436 -> 607.312 -> 5465.812 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 634.774 501.303 554.786 -> 634.774 -> 5712.969 MByte/s p17 best bi-section : 359.987 273.558 507.293 -> 507.293 -> 4565.633 MByte/s p18 worst bi-section : 360.742 213.700 507.642 -> 507.642 -> 4568.776 MByte/s p19 acyclic-1dim-all : 561.929 446.179 500.938 -> 561.929 -> 5057.360 MByte/s p20 acyclic-2dim-all : 419.834 420.192 400.251 -> 420.192 -> 3781.724 MByte/s p21 acyclic-3dim-all : 361.056 390.891 436.135 -> 436.135 -> 3925.213 MByte/s p22 cyclic-1dim-all : 650.601 501.276 562.319 -> 650.601 -> 5855.405 MByte/s p23 cyclic-2dim-all : 616.032 490.326 488.662 -> 616.032 -> 5544.284 MByte/s p24 cyclic-3dim-all : 579.685 390.909 461.040 -> 579.685 -> 5217.167 MByte/s log_avg of all rings : 637.246 391.460 542.289 || 637.246 -> 5735.212 MByte/s log_avg of all random : 634.062 499.543 545.339 || 634.062 -> 5706.557 MByte/s log_avg(ring,random) : 635.652 442.211 543.812 ||(635.652 -> 5720.867)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-4*2&+1 : 628.281 646.205 645.634 -> 646.205 -> 5815.847 MByte/s p01 ring-2*4&+1 : 586.049 570.170 612.293 -> 612.293 -> 5510.640 MByte/s p02 ring-1*9fix : 644.094 605.721 627.252 -> 644.094 -> 5796.842 MByte/s p03 ring-1*9fix : 644.355 621.088 630.617 -> 644.355 -> 5799.194 MByte/s p04 ring-1*9fix : 627.619 610.648 623.369 -> 627.619 -> 5648.568 MByte/s p05 ring-1*9fix : 643.781 609.375 635.502 -> 643.781 -> 5794.026 MByte/s p06 random-cyc-1dim : 621.353 619.257 630.872 -> 630.872 -> 5677.852 MByte/s p07 random-cyc-1dim : 623.754 614.053 582.631 -> 623.754 -> 5613.784 MByte/s p08 random-cyc-1dim : 626.011 616.732 632.404 -> 632.404 -> 5691.633 MByte/s p09 random-cyc-1dim : 627.338 616.084 642.275 -> 642.275 -> 5780.474 MByte/s p10 random-cyc-1dim : 638.317 624.275 644.054 -> 644.054 -> 5796.488 MByte/s p11 random-cyc-1dim : 617.825 621.382 638.333 -> 638.333 -> 5744.999 MByte/s p12 random-cyc-1dim : 619.736 620.595 637.932 -> 637.932 -> 5741.385 MByte/s p13 random-cyc-1dim : 611.962 615.383 623.026 -> 623.026 -> 5607.231 MByte/s p14 random-cyc-1dim : 621.931 624.394 595.861 -> 624.394 -> 5619.547 MByte/s p15 random-cyc-1dim : 621.143 596.112 624.045 -> 624.045 -> 5616.407 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 618.105 625.425 641.257 -> 641.257 -> 5771.317 MByte/s p17 best bi-section : 503.011 508.624 465.222 -> 508.624 -> 4577.612 MByte/s p18 worst bi-section : 507.722 503.588 506.802 -> 507.722 -> 4569.500 MByte/s p19 acyclic-1dim-all : 548.819 555.014 564.397 -> 564.397 -> 5079.576 MByte/s p20 acyclic-2dim-all : 436.351 429.583 436.054 -> 436.351 -> 3927.155 MByte/s p21 acyclic-3dim-all : 417.807 411.654 437.102 -> 437.102 -> 3933.921 MByte/s p22 cyclic-1dim-all : 610.362 630.058 648.782 -> 648.782 -> 5839.034 MByte/s p23 cyclic-2dim-all : 587.546 589.818 583.492 -> 589.818 -> 5308.361 MByte/s p24 cyclic-3dim-all : 572.396 550.176 574.861 -> 574.861 -> 5173.749 MByte/s log_avg of all rings : 628.684 610.116 629.027 || 636.267 -> 5726.406 MByte/s log_avg of all random : 622.903 616.778 624.836 || 632.062 -> 5688.555 MByte/s log_avg(ring,random) : 625.787 613.438 626.928 ||(634.161 -> 5707.450)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-4*2&+1 p00 method 0 : 0.051 0.846 13.323 163.561 1453.640 3514.889 -> 648.224 -> 5834.018 MByte/s p00 method 1 : 0.008 0.145 2.282 36.594 423.184 869.039 -> 178.427 -> 1605.847 MByte/s p00 method 2 : 0.035 0.572 8.965 124.007 1215.192 2647.585 -> 526.151 -> 4735.358 MByte/s p01 ring-2*4&+1 p01 method 0 : 0.051 0.851 13.385 157.348 1454.665 2876.057 -> 617.135 -> 5554.213 MByte/s p01 method 1 : 0.015 0.288 4.519 71.820 822.401 1444.821 -> 325.034 -> 2925.303 MByte/s p01 method 2 : 0.040 0.666 10.427 139.023 1362.501 2429.993 -> 522.537 -> 4702.830 MByte/s p02 ring-1*9fix p02 method 0 : 0.051 0.851 13.406 159.369 1451.294 3577.610 -> 647.602 -> 5828.419 MByte/s p02 method 1 : 0.014 0.288 4.552 71.954 823.684 3206.615 -> 497.482 -> 4477.337 MByte/s p02 method 2 : 0.037 0.664 10.382 134.827 1363.787 2618.089 -> 544.032 -> 4896.290 MByte/s p03 ring-1*9fix p03 method 0 : 0.051 0.850 13.401 162.797 1445.319 3566.221 -> 649.015 -> 5841.131 MByte/s p03 method 1 : 0.015 0.285 4.549 71.941 827.008 3230.981 -> 499.821 -> 4498.392 MByte/s p03 method 2 : 0.040 0.663 10.358 136.036 1358.314 2857.996 -> 559.231 -> 5033.077 MByte/s p04 ring-1*9fix p04 method 0 : 0.052 0.850 13.391 163.475 1448.984 2872.638 -> 616.015 -> 5544.134 MByte/s p04 method 1 : 0.015 0.287 4.550 72.136 825.625 3219.869 -> 499.748 -> 4497.728 MByte/s p04 method 2 : 0.040 0.662 10.395 134.598 1362.104 2600.880 -> 539.043 -> 4851.383 MByte/s p05 ring-1*9fix p05 method 0 : 0.051 0.850 13.379 163.609 1457.489 3572.588 -> 646.521 -> 5818.687 MByte/s p05 method 1 : 0.015 0.285 4.550 71.772 824.282 3221.155 -> 499.333 -> 4493.996 MByte/s p05 method 2 : 0.040 0.666 10.368 134.609 1361.871 2827.631 -> 564.055 -> 5076.493 MByte/s p06 random-cyc-1dim p06 method 0 : 0.051 0.851 13.398 164.010 1449.284 3538.542 -> 631.226 -> 5681.035 MByte/s p06 method 1 : 0.015 0.287 4.549 71.777 825.659 3197.502 -> 499.761 -> 4497.853 MByte/s p06 method 2 : 0.040 0.668 10.392 135.869 1362.271 2622.056 -> 539.920 -> 4859.283 MByte/s p07 random-cyc-1dim p07 method 0 : 0.050 0.817 12.904 159.162 1447.654 3545.721 -> 635.338 -> 5718.043 MByte/s p07 method 1 : 0.015 0.288 4.520 71.984 824.362 3190.361 -> 496.847 -> 4471.620 MByte/s p07 method 2 : 0.039 0.657 10.281 136.285 1362.679 2819.966 -> 557.611 -> 5018.502 MByte/s p08 random-cyc-1dim p08 method 0 : 0.049 0.818 12.898 158.302 1445.282 3548.841 -> 627.502 -> 5647.521 MByte/s p08 method 1 : 0.015 0.288 4.539 71.953 820.864 3213.357 -> 499.442 -> 4494.981 MByte/s p08 method 2 : 0.040 0.659 10.266 137.320 1362.544 2399.696 -> 520.268 -> 4682.408 MByte/s p09 random-cyc-1dim p09 method 0 : 0.052 0.851 13.398 162.118 1444.160 3532.343 -> 645.864 -> 5812.775 MByte/s p09 method 1 : 0.015 0.288 4.552 71.920 825.321 3203.049 -> 498.845 -> 4489.606 MByte/s p09 method 2 : 0.040 0.666 10.388 136.910 1362.493 2652.407 -> 538.408 -> 4845.668 MByte/s p10 random-cyc-1dim p10 method 0 : 0.051 0.851 13.395 161.822 1442.375 3560.965 -> 646.788 -> 5821.093 MByte/s p10 method 1 : 0.015 0.288 4.556 71.901 820.994 3196.820 -> 499.041 -> 4491.366 MByte/s p10 method 2 : 0.040 0.664 10.410 138.834 1363.860 2963.603 -> 565.664 -> 5090.977 MByte/s p11 random-cyc-1dim p11 method 0 : 0.052 0.850 13.419 157.550 1450.051 3534.724 -> 629.655 -> 5666.892 MByte/s p11 method 1 : 0.015 0.288 4.553 71.869 830.151 3220.858 -> 500.639 -> 4505.755 MByte/s p11 method 2 : 0.040 0.666 10.400 136.486 1365.574 2887.971 -> 549.748 -> 4947.734 MByte/s p12 random-cyc-1dim p12 method 0 : 0.051 0.850 13.394 163.223 1448.712 3569.305 -> 634.030 -> 5706.270 MByte/s p12 method 1 : 0.015 0.288 4.549 71.889 822.835 3192.945 -> 499.457 -> 4495.110 MByte/s p12 method 2 : 0.040 0.665 10.411 136.613 1363.054 2877.604 -> 557.472 -> 5017.250 MByte/s p13 random-cyc-1dim p13 method 0 : 0.050 0.820 12.936 159.489 1443.212 3569.256 -> 647.046 -> 5823.411 MByte/s p13 method 1 : 0.015 0.288 4.549 72.057 826.943 3200.293 -> 500.713 -> 4506.414 MByte/s p13 method 2 : 0.040 0.667 10.391 137.831 1364.101 2417.355 -> 528.461 -> 4756.150 MByte/s p14 random-cyc-1dim p14 method 0 : 0.050 0.820 12.934 159.535 1444.565 3560.434 -> 636.867 -> 5731.805 MByte/s p14 method 1 : 0.015 0.285 4.505 71.351 823.114 3184.916 -> 498.465 -> 4486.188 MByte/s p14 method 2 : 0.039 0.652 9.892 135.284 1354.262 2879.295 -> 554.067 -> 4986.602 MByte/s p15 random-cyc-1dim p15 method 0 : 0.051 0.850 13.391 163.754 1454.740 2817.586 -> 607.312 -> 5465.812 MByte/s p15 method 1 : 0.015 0.288 4.536 72.108 821.865 3218.683 -> 502.239 -> 4520.149 MByte/s p15 method 2 : 0.040 0.667 10.407 137.391 1358.941 2868.724 -> 543.436 -> 4890.921 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.051 0.850 13.411 163.622 1447.367 3573.245 -> 634.774 -> 5712.969 MByte/s p16 method 1 : 0.015 0.288 4.562 71.964 828.640 3207.027 -> 501.303 -> 4511.725 MByte/s p16 method 2 : 0.040 0.667 10.426 137.155 1363.492 2862.662 -> 554.786 -> 4993.076 MByte/s p17 best bi-section p17 method 0 : 0.035 0.579 9.003 103.590 747.170 2165.464 -> 359.987 -> 3239.879 MByte/s p17 method 1 : 0.007 0.129 2.051 32.591 458.761 1429.578 -> 273.558 -> 2462.022 MByte/s p17 method 2 : 0.024 0.392 6.124 86.923 966.664 3091.762 -> 507.293 -> 4565.633 MByte/s p18 worst bi-section p18 method 0 : 0.035 0.578 8.999 103.562 745.800 2171.437 -> 360.742 -> 3246.678 MByte/s p18 method 1 : 0.007 0.129 2.042 32.571 457.011 647.717 -> 213.700 -> 1923.302 MByte/s p18 method 2 : 0.024 0.392 6.129 87.267 967.060 3091.393 -> 507.642 -> 4568.776 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.046 0.757 11.921 145.697 1301.519 3168.703 -> 561.929 -> 5057.360 MByte/s p19 method 1 : 0.013 0.255 4.040 63.913 746.345 2840.820 -> 446.179 -> 4015.611 MByte/s p19 method 2 : 0.033 0.576 9.043 119.159 1200.624 2512.623 -> 500.938 -> 4508.438 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.034 0.570 8.959 109.853 978.314 2156.166 -> 419.834 -> 3778.503 MByte/s p20 method 1 : 0.019 0.379 5.943 93.872 836.754 2283.203 -> 420.192 -> 3781.724 MByte/s p20 method 2 : 0.027 0.449 7.094 94.965 904.474 2096.620 -> 400.251 -> 3602.258 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.036 0.589 9.174 104.698 751.777 2152.276 -> 361.056 -> 3249.505 MByte/s p21 method 1 : 0.020 0.419 6.526 103.034 983.777 1896.903 -> 390.891 -> 3518.016 MByte/s p21 method 2 : 0.024 0.399 6.114 89.186 1033.558 2343.220 -> 436.135 -> 3925.213 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.051 0.851 13.396 164.197 1466.542 3558.065 -> 650.601 -> 5855.405 MByte/s p22 method 1 : 0.015 0.288 4.542 72.059 827.901 3216.333 -> 501.276 -> 4511.482 MByte/s p22 method 2 : 0.039 0.652 10.085 133.921 1360.181 2805.104 -> 562.319 -> 5060.872 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.051 0.854 13.440 164.528 1465.130 2845.726 -> 616.032 -> 5544.284 MByte/s p23 method 1 : 0.027 0.564 8.853 138.164 1171.506 2371.271 -> 490.326 -> 4412.935 MByte/s p23 method 2 : 0.035 0.582 10.425 122.211 1287.247 2136.607 -> 488.662 -> 4397.956 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.045 0.741 11.673 143.462 1308.554 3170.815 -> 579.685 -> 5217.167 MByte/s p24 method 1 : 0.020 0.422 6.647 103.798 974.207 1894.415 -> 390.909 -> 3518.184 MByte/s p24 method 2 : 0.032 0.526 8.540 116.127 1186.014 2353.857 -> 461.040 -> 4149.363 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.850 13.381 161.674 1451.893 3313.567 || 637.246 -> 5735.212 MByte/s - ring, method 1 : 0.013 0.256 4.051 64.264 737.831 2264.718 || 391.460 -> 3523.138 MByte/s - ring, method 2 : 0.039 0.648 10.134 133.766 1336.121 2659.749 || 542.289 -> 4880.605 MByte/s log_avg of all random - random, method 0 : 0.051 0.838 13.205 160.881 1446.999 3469.879 || 634.062 -> 5706.557 MByte/s - random, method 1 : 0.015 0.287 4.541 71.881 824.206 3201.858 || 499.543 -> 4495.887 MByte/s - random, method 2 : 0.040 0.663 10.323 136.879 1361.975 2731.770 || 545.339 -> 4908.054 MByte/s log_avg(ring,random) - average, method 0 : 0.051 0.844 13.292 161.277 1449.444 3390.822 || 635.652 -> 5720.867 MByte/s - average, method 1 : 0.014 0.271 4.289 67.966 779.824 2692.825 || 442.211 -> 3979.903 MByte/s - average, method 2 : 0.039 0.655 10.228 135.313 1348.986 2695.519 || 543.812 -> 4894.310 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 0.458 0.051 0.051 0.051 0.051 0.014 0.039 2 0.905 0.101 0.101 0.100 0.101 0.028 0.078 4 1.861 0.207 0.208 0.205 0.207 0.068 0.161 8 3.792 0.421 0.424 0.419 0.421 0.136 0.327 16 7.593 0.844 0.850 0.838 0.844 0.271 0.655 32 15.141 1.682 1.694 1.671 1.682 0.542 1.305 64 30.169 3.352 3.376 3.328 3.352 1.081 2.593 128 59.860 6.651 6.697 6.606 6.651 2.139 5.116 256 119.632 13.292 13.381 13.205 13.292 4.289 10.228 512 239.209 26.579 26.749 26.409 26.579 8.593 20.498 1024 476.464 52.940 53.303 52.581 52.940 17.191 40.793 2048 732.124 81.347 81.748 80.948 81.347 34.325 68.953 4096 1451.494 161.277 161.674 160.881 161.277 67.966 135.313 8933 1903.422 211.491 211.924 211.060 211.491 118.085 187.658 19484 5004.751 556.083 556.617 555.550 556.083 234.940 496.367 42495 7446.170 827.352 828.440 826.266 827.352 422.189 757.819 92682 13044.996 1449.444 1451.893 1446.999 1449.444 779.824 1348.986 202141 15189.601 1687.733 1688.777 1686.691 1687.733 1271.240 1524.309 440872 25231.429 2803.492 2808.346 2798.647 2803.492 1853.142 2429.613 961548 19046.087 2116.232 2159.250 2074.071 2057.962 1715.219 1687.082 2097152 31014.682 3446.076 3377.188 3516.369 3390.822 2692.825 2695.519 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-4*2&+1 : 0.051 0.846 13.323 163.561 1453.640 3514.889 -> 648.224 -> 5834.018 MByte/s p01 ring-2*4&+1 : 0.051 0.851 13.385 157.348 1454.665 2876.057 -> 617.135 -> 5554.213 MByte/s p02 ring-1*9fix : 0.051 0.851 13.406 159.369 1451.294 3577.610 -> 647.602 -> 5828.419 MByte/s p03 ring-1*9fix : 0.051 0.850 13.401 162.797 1445.319 3566.221 -> 649.015 -> 5841.131 MByte/s p04 ring-1*9fix : 0.052 0.850 13.391 163.475 1448.984 3219.869 -> 632.550 -> 5692.948 MByte/s p05 ring-1*9fix : 0.051 0.850 13.379 163.609 1457.489 3572.588 -> 646.521 -> 5818.687 MByte/s p06 random-cyc-1dim : 0.051 0.851 13.398 164.010 1449.284 3538.542 -> 640.487 -> 5764.381 MByte/s p07 random-cyc-1dim : 0.050 0.817 12.904 159.162 1447.654 3545.721 -> 637.039 -> 5733.353 MByte/s p08 random-cyc-1dim : 0.049 0.818 12.898 158.302 1445.282 3548.841 -> 638.851 -> 5749.663 MByte/s p09 random-cyc-1dim : 0.052 0.851 13.398 162.118 1444.160 3532.343 -> 645.864 -> 5812.775 MByte/s p10 random-cyc-1dim : 0.051 0.851 13.395 161.822 1442.375 3560.965 -> 646.788 -> 5821.093 MByte/s p11 random-cyc-1dim : 0.052 0.850 13.419 157.550 1450.051 3534.724 -> 641.565 -> 5774.089 MByte/s p12 random-cyc-1dim : 0.051 0.850 13.394 163.223 1448.712 3569.305 -> 643.312 -> 5789.811 MByte/s p13 random-cyc-1dim : 0.050 0.820 12.936 159.489 1443.212 3569.256 -> 647.046 -> 5823.411 MByte/s p14 random-cyc-1dim : 0.050 0.820 12.934 159.535 1444.565 3560.434 -> 642.037 -> 5778.337 MByte/s p15 random-cyc-1dim : 0.051 0.850 13.391 163.754 1454.740 3218.683 -> 629.166 -> 5662.498 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.051 0.850 13.411 163.622 1447.367 3573.245 -> 645.577 -> 5810.197 MByte/s p17 best bi-section : 0.035 0.579 9.003 103.590 966.664 3091.762 -> 509.600 -> 4586.403 MByte/s p18 worst bi-section : 0.035 0.578 8.999 103.562 967.060 3091.393 -> 509.914 -> 4589.228 MByte/s p19 acyclic-1dim-all : 0.046 0.757 11.921 145.697 1301.519 3168.703 -> 572.489 -> 5152.402 MByte/s p20 acyclic-2dim-all : 0.034 0.570 8.959 109.853 978.314 2283.203 -> 438.680 -> 3948.116 MByte/s p21 acyclic-3dim-all : 0.036 0.589 9.174 104.698 1033.558 2343.220 -> 439.568 -> 3956.114 MByte/s p22 cyclic-1dim-all : 0.051 0.851 13.396 164.197 1466.542 3558.065 -> 650.601 -> 5855.405 MByte/s p23 cyclic-2dim-all : 0.051 0.854 13.440 164.528 1465.130 2845.726 -> 616.032 -> 5544.284 MByte/s p24 cyclic-3dim-all : 0.045 0.741 11.673 143.462 1308.554 3170.815 -> 579.685 -> 5217.167 MByte/s log_avg of all rings : 0.051 0.850 13.381 161.674 1451.893 3377.188 || 640.065 -> 5760.587 MByte/s log_avg of all random : 0.051 0.838 13.205 160.881 1446.999 3516.369 || 641.195 -> 5770.757 MByte/s log_avg(ring,random) : 0.051 0.844 13.292 161.277 1449.444 3446.076 || 640.630 -> 5765.670 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 5765.670 MByte/s on 9 processes ( = 640.630 MByte/s * 9 processes) system parameters : 9 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 = 5765.670 MB/s = 640.630 * 9 PEs with 256 MB/PE on SUPER-UX hwwsx4 9.1 Rev1 SX-4