b_eff = 9007.233 MB/s = 643.374 * 14 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 = 14 1-dim-paterns: size = 14 2-dim-paterns: size = 7 * 2 3-dim-paterns: size = 3 * 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 = 86.437 sec sum of max elapsed time per entries above = 86.268 sec difference = 0.169 sec = 0.2% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-7*2fix => 1 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p01 ring-3*4&+1 => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p02 ring-2*7fix => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p03 ring-1*14fix => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p04 ring-1*14fix => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p05 ring-1*14fix => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 26 messages, i.e. 4.2 msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 38 messages, i.e. 4.2 msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 40 messages, i.e. 4.2 msgs/used node, 2 nodes are UNUSED p22 cyclic-1dim-all => 2 sendrecv_calls with 28 messages, i.e. 4.2 msgs/used node, all nodes are used p23 cyclic-2dim-all => 3 sendrecv_calls with 42 messages, i.e. 4.2 msgs/used node, all nodes are used p24 cyclic-3dim-all => 4 sendrecv_calls with 48 messages, i.e. 4.2 msgs/used node, 2 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-7*2fix : 650.153 282.095 598.247 -> 650.153 -> 9102.148 MByte/s p01 ring-3*4&+1 : 652.094 304.088 515.769 -> 652.094 -> 9129.309 MByte/s p02 ring-2*7fix : 637.351 329.011 551.162 -> 637.351 -> 8922.914 MByte/s p03 ring-1*14fix : 633.414 457.156 549.657 -> 633.414 -> 8867.802 MByte/s p04 ring-1*14fix : 632.336 456.894 550.310 -> 632.336 -> 8852.702 MByte/s p05 ring-1*14fix : 649.291 457.814 557.062 -> 649.291 -> 9090.076 MByte/s p06 random-cyc-1dim : 639.075 455.910 523.457 -> 639.075 -> 8947.044 MByte/s p07 random-cyc-1dim : 649.846 457.669 508.904 -> 649.846 -> 9097.846 MByte/s p08 random-cyc-1dim : 647.592 456.440 528.334 -> 647.592 -> 9066.281 MByte/s p09 random-cyc-1dim : 648.295 457.181 566.372 -> 648.295 -> 9076.127 MByte/s p10 random-cyc-1dim : 624.941 455.673 527.358 -> 624.941 -> 8749.169 MByte/s p11 random-cyc-1dim : 648.360 456.950 559.581 -> 648.360 -> 9077.035 MByte/s p12 random-cyc-1dim : 649.891 459.041 517.670 -> 649.891 -> 9098.467 MByte/s p13 random-cyc-1dim : 615.196 457.285 523.254 -> 615.196 -> 8612.739 MByte/s p14 random-cyc-1dim : 647.855 456.210 552.038 -> 647.855 -> 9069.972 MByte/s p15 random-cyc-1dim : 649.780 457.767 540.759 -> 649.780 -> 9096.925 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 648.573 458.247 529.412 -> 648.573 -> 9080.024 MByte/s p17 best bi-section : 405.269 280.682 565.549 -> 565.549 -> 7917.690 MByte/s p18 worst bi-section : 405.810 202.870 566.027 -> 566.027 -> 7924.373 MByte/s p19 acyclic-1dim-all : 592.143 426.768 524.735 -> 592.143 -> 8290.000 MByte/s p20 acyclic-2dim-all : 474.306 404.995 513.902 -> 513.902 -> 7194.630 MByte/s p21 acyclic-3dim-all : 353.877 339.183 407.513 -> 407.513 -> 5705.181 MByte/s p22 cyclic-1dim-all : 600.286 457.209 551.611 -> 600.286 -> 8404.009 MByte/s p23 cyclic-2dim-all : 641.559 447.086 550.256 -> 641.559 -> 8981.826 MByte/s p24 cyclic-3dim-all : 547.356 404.143 430.064 -> 547.356 -> 7662.978 MByte/s log_avg of all rings : 642.387 373.132 553.184 || 642.387 -> 8993.415 MByte/s log_avg of all random : 641.976 457.012 534.470 || 641.976 -> 8987.665 MByte/s log_avg(ring,random) : 642.181 412.947 543.746 ||(642.181 -> 8990.540)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-7*2fix : 635.868 648.630 642.008 -> 648.630 -> 9080.813 MByte/s p01 ring-3*4&+1 : 589.131 617.511 633.776 -> 633.776 -> 8872.867 MByte/s p02 ring-2*7fix : 577.917 588.382 628.654 -> 628.654 -> 8801.157 MByte/s p03 ring-1*14fix : 605.181 614.569 633.587 -> 633.587 -> 8870.212 MByte/s p04 ring-1*14fix : 606.504 608.279 625.433 -> 625.433 -> 8756.058 MByte/s p05 ring-1*14fix : 630.758 610.134 623.732 -> 630.758 -> 8830.612 MByte/s p06 random-cyc-1dim : 613.010 637.492 633.090 -> 637.492 -> 8924.887 MByte/s p07 random-cyc-1dim : 629.430 609.995 635.567 -> 635.567 -> 8897.943 MByte/s p08 random-cyc-1dim : 613.679 620.711 633.606 -> 633.606 -> 8870.477 MByte/s p09 random-cyc-1dim : 600.750 619.824 640.291 -> 640.291 -> 8964.072 MByte/s p10 random-cyc-1dim : 628.203 604.639 607.698 -> 628.203 -> 8794.836 MByte/s p11 random-cyc-1dim : 610.916 608.364 647.163 -> 647.163 -> 9060.288 MByte/s p12 random-cyc-1dim : 642.476 608.586 648.694 -> 648.694 -> 9081.721 MByte/s p13 random-cyc-1dim : 604.759 610.710 608.947 -> 610.710 -> 8549.940 MByte/s p14 random-cyc-1dim : 643.173 618.152 622.600 -> 643.173 -> 9004.422 MByte/s p15 random-cyc-1dim : 620.667 641.935 623.591 -> 641.935 -> 8987.097 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 632.500 615.620 646.271 -> 646.271 -> 9047.793 MByte/s p17 best bi-section : 559.619 567.243 568.077 -> 568.077 -> 7953.077 MByte/s p18 worst bi-section : 558.193 564.950 568.738 -> 568.738 -> 7962.327 MByte/s p19 acyclic-1dim-all : 569.006 581.183 598.323 -> 598.323 -> 8376.522 MByte/s p20 acyclic-2dim-all : 487.772 506.220 515.538 -> 515.538 -> 7217.535 MByte/s p21 acyclic-3dim-all : 397.363 409.583 402.761 -> 409.583 -> 5734.159 MByte/s p22 cyclic-1dim-all : 606.857 601.881 609.744 -> 609.744 -> 8536.413 MByte/s p23 cyclic-2dim-all : 620.134 604.747 594.787 -> 620.134 -> 8681.878 MByte/s p24 cyclic-3dim-all : 535.154 539.404 522.474 -> 539.404 -> 7551.661 MByte/s log_avg of all rings : 607.208 614.327 631.169 || 633.430 -> 8868.027 MByte/s log_avg of all random : 620.548 617.926 629.978 || 636.596 -> 8912.347 MByte/s log_avg(ring,random) : 613.842 616.124 630.573 ||(635.011 -> 8890.159)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-7*2fix p00 method 0 : 0.051 0.846 13.330 163.903 1459.576 3533.605 -> 650.153 -> 9102.148 MByte/s p00 method 1 : 0.006 0.120 1.900 30.503 452.234 1539.723 -> 282.095 -> 3949.324 MByte/s p00 method 2 : 0.035 0.542 8.502 122.978 1211.080 3498.425 -> 598.247 -> 8375.453 MByte/s p01 ring-3*4&+1 p01 method 0 : 0.051 0.852 13.409 164.623 1465.606 3544.762 -> 652.094 -> 9129.309 MByte/s p01 method 1 : 0.012 0.238 3.744 60.091 743.129 1417.585 -> 304.088 -> 4257.238 MByte/s p01 method 2 : 0.040 0.629 9.999 135.831 1354.640 2377.497 -> 515.769 -> 7220.765 MByte/s p02 ring-2*7fix p02 method 0 : 0.052 0.852 13.411 158.396 1461.790 3537.778 -> 637.351 -> 8922.914 MByte/s p02 method 1 : 0.012 0.238 3.756 60.134 742.410 1695.129 -> 329.011 -> 4606.157 MByte/s p02 method 2 : 0.040 0.632 9.975 135.872 1358.695 2867.610 -> 551.162 -> 7716.271 MByte/s p03 ring-1*14fix p03 method 0 : 0.052 0.851 13.414 161.375 1454.055 3555.845 -> 633.414 -> 8867.802 MByte/s p03 method 1 : 0.012 0.238 3.758 60.134 745.999 3053.850 -> 457.156 -> 6400.189 MByte/s p03 method 2 : 0.040 0.630 9.985 134.205 1354.143 2830.317 -> 549.657 -> 7695.199 MByte/s p04 ring-1*14fix p04 method 0 : 0.052 0.851 13.406 165.093 1452.874 3543.491 -> 632.336 -> 8852.702 MByte/s p04 method 1 : 0.012 0.238 3.773 60.066 743.018 3047.141 -> 456.894 -> 6396.510 MByte/s p04 method 2 : 0.039 0.630 9.993 134.655 1354.309 2771.107 -> 550.310 -> 7704.336 MByte/s p05 ring-1*14fix p05 method 0 : 0.052 0.850 13.409 165.181 1451.964 3548.073 -> 649.291 -> 9090.076 MByte/s p05 method 1 : 0.012 0.238 3.765 59.945 743.557 3044.416 -> 457.814 -> 6409.389 MByte/s p05 method 2 : 0.040 0.629 9.985 135.230 1355.613 2800.669 -> 557.062 -> 7798.870 MByte/s p06 random-cyc-1dim p06 method 0 : 0.051 0.851 13.396 164.565 1452.365 3533.176 -> 639.075 -> 8947.044 MByte/s p06 method 1 : 0.012 0.238 3.774 60.122 742.178 3030.864 -> 455.910 -> 6382.736 MByte/s p06 method 2 : 0.038 0.642 9.985 135.250 1359.942 2571.211 -> 523.457 -> 7328.392 MByte/s p07 random-cyc-1dim p07 method 0 : 0.051 0.852 13.404 164.165 1452.191 3568.455 -> 649.846 -> 9097.846 MByte/s p07 method 1 : 0.012 0.238 3.775 60.026 747.444 3050.935 -> 457.669 -> 6407.372 MByte/s p07 method 2 : 0.039 0.617 9.995 133.064 1356.941 2359.235 -> 508.904 -> 7124.654 MByte/s p08 random-cyc-1dim p08 method 0 : 0.051 0.852 13.407 164.911 1445.774 3557.752 -> 647.592 -> 9066.281 MByte/s p08 method 1 : 0.012 0.238 3.769 60.000 740.439 3053.850 -> 456.440 -> 6390.154 MByte/s p08 method 2 : 0.040 0.624 9.998 132.774 1357.490 2560.700 -> 528.334 -> 7396.671 MByte/s p09 random-cyc-1dim p09 method 0 : 0.052 0.851 13.410 164.734 1450.328 3539.736 -> 648.295 -> 9076.127 MByte/s p09 method 1 : 0.012 0.238 3.771 60.138 742.532 3048.381 -> 457.181 -> 6400.534 MByte/s p09 method 2 : 0.040 0.625 9.989 133.466 1354.820 3174.579 -> 566.372 -> 7929.210 MByte/s p10 random-cyc-1dim p10 method 0 : 0.051 0.852 13.416 159.559 1451.323 3487.071 -> 624.941 -> 8749.169 MByte/s p10 method 1 : 0.012 0.238 3.775 60.043 739.080 3024.587 -> 455.673 -> 6379.420 MByte/s p10 method 2 : 0.040 0.646 10.008 134.549 1348.498 2574.747 -> 527.358 -> 7383.006 MByte/s p11 random-cyc-1dim p11 method 0 : 0.052 0.852 13.415 163.557 1449.672 3531.130 -> 648.360 -> 9077.035 MByte/s p11 method 1 : 0.012 0.238 3.778 60.139 744.855 3045.441 -> 456.950 -> 6397.306 MByte/s p11 method 2 : 0.040 0.630 10.000 134.306 1356.042 3107.868 -> 559.581 -> 7834.135 MByte/s p12 random-cyc-1dim p12 method 0 : 0.052 0.851 13.408 164.695 1448.103 3542.821 -> 649.891 -> 9098.467 MByte/s p12 method 1 : 0.012 0.238 3.775 60.114 743.299 3065.994 -> 459.041 -> 6426.568 MByte/s p12 method 2 : 0.040 0.648 9.989 135.520 1360.250 2529.468 -> 517.670 -> 7247.384 MByte/s p13 random-cyc-1dim p13 method 0 : 0.052 0.851 13.408 162.601 1445.729 2853.283 -> 615.196 -> 8612.739 MByte/s p13 method 1 : 0.012 0.237 3.760 59.914 742.547 3040.655 -> 457.285 -> 6401.996 MByte/s p13 method 2 : 0.040 0.646 10.102 133.422 1356.565 2600.996 -> 523.254 -> 7325.562 MByte/s p14 random-cyc-1dim p14 method 0 : 0.052 0.851 13.416 164.422 1441.143 3514.913 -> 647.855 -> 9069.972 MByte/s p14 method 1 : 0.012 0.238 3.780 60.051 741.251 3026.613 -> 456.210 -> 6386.938 MByte/s p14 method 2 : 0.040 0.627 10.019 134.060 1356.150 2835.062 -> 552.038 -> 7728.529 MByte/s p15 random-cyc-1dim p15 method 0 : 0.051 0.852 13.412 164.713 1449.272 3536.752 -> 649.780 -> 9096.925 MByte/s p15 method 1 : 0.012 0.238 3.780 60.130 739.536 3040.373 -> 457.767 -> 6408.745 MByte/s p15 method 2 : 0.040 0.629 10.004 132.626 1356.919 2773.981 -> 540.759 -> 7570.622 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.051 0.851 13.423 164.511 1448.658 3547.352 -> 648.573 -> 9080.024 MByte/s p16 method 1 : 0.012 0.238 3.780 60.133 739.982 3045.460 -> 458.247 -> 6415.451 MByte/s p16 method 2 : 0.040 0.644 10.005 134.272 1356.565 2601.861 -> 529.412 -> 7411.767 MByte/s p17 best bi-section p17 method 0 : 0.039 0.649 10.134 116.810 839.577 2426.495 -> 405.269 -> 5673.766 MByte/s p17 method 1 : 0.006 0.120 1.901 30.381 448.072 1530.580 -> 280.682 -> 3929.541 MByte/s p17 method 2 : 0.028 0.427 6.690 97.104 1081.502 3427.839 -> 565.549 -> 7917.690 MByte/s p18 worst bi-section p18 method 0 : 0.039 0.649 10.131 116.685 840.319 2429.486 -> 405.810 -> 5681.344 MByte/s p18 method 1 : 0.006 0.120 1.902 30.321 449.736 512.982 -> 202.870 -> 2840.182 MByte/s p18 method 2 : 0.027 0.427 6.688 97.167 1082.403 3429.139 -> 566.027 -> 7924.373 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.048 0.790 12.457 153.285 1357.739 3292.811 -> 592.143 -> 8290.000 MByte/s p19 method 1 : 0.012 0.221 3.507 55.661 694.478 2824.202 -> 426.768 -> 5974.759 MByte/s p19 method 2 : 0.034 0.545 9.209 125.510 1256.351 2629.784 -> 524.735 -> 7346.289 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.039 0.650 10.205 124.801 1044.414 2776.706 -> 474.306 -> 6640.289 MByte/s p20 method 1 : 0.016 0.319 5.029 79.747 868.529 2266.803 -> 404.995 -> 5669.937 MByte/s p20 method 2 : 0.031 0.504 7.826 110.108 1195.393 2756.590 -> 513.902 -> 7194.630 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.030 0.502 7.883 94.667 758.282 2068.625 -> 353.877 -> 4954.278 MByte/s p21 method 1 : 0.018 0.357 5.606 88.925 828.102 1678.182 -> 339.183 -> 4748.560 MByte/s p21 method 2 : 0.025 0.409 6.376 89.059 946.058 2010.520 -> 407.513 -> 5705.181 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.051 0.850 13.411 164.902 1456.498 2848.787 -> 600.286 -> 8404.009 MByte/s p22 method 1 : 0.012 0.237 3.776 60.052 743.128 3028.168 -> 457.209 -> 6400.927 MByte/s p22 method 2 : 0.038 0.624 9.975 134.756 1359.307 2866.983 -> 551.611 -> 7722.559 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.051 0.853 13.449 164.714 1457.356 3496.870 -> 641.559 -> 8981.826 MByte/s p23 method 1 : 0.018 0.354 5.607 88.799 946.825 2508.179 -> 447.086 -> 6259.200 MByte/s p23 method 2 : 0.038 0.646 10.206 136.489 1461.671 2764.124 -> 550.256 -> 7703.579 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.044 0.732 11.545 141.806 1260.703 3044.646 -> 547.356 -> 7662.978 MByte/s p24 method 1 : 0.021 0.432 6.826 107.645 971.980 2012.308 -> 404.143 -> 5658.007 MByte/s p24 method 2 : 0.030 0.469 6.983 101.689 1114.403 2078.021 -> 430.064 -> 6020.891 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.850 13.396 163.077 1457.635 3543.919 || 642.387 -> 8993.415 MByte/s - ring, method 1 : 0.011 0.212 3.355 53.657 684.469 2171.400 || 373.132 -> 5223.846 MByte/s - ring, method 2 : 0.039 0.614 9.723 133.046 1330.269 2839.295 || 553.184 -> 7744.576 MByte/s log_avg of all random - random, method 0 : 0.051 0.851 13.409 163.785 1448.586 3459.696 || 641.976 -> 8987.665 MByte/s - random, method 1 : 0.012 0.238 3.774 60.068 742.312 3042.745 || 457.012 -> 6398.163 MByte/s - random, method 2 : 0.040 0.633 10.009 133.900 1356.358 2697.731 || 534.470 -> 7482.580 MByte/s log_avg(ring,random) - average, method 0 : 0.051 0.851 13.403 163.430 1453.104 3501.554 || 642.181 -> 8990.540 MByte/s - average, method 1 : 0.012 0.225 3.558 56.772 712.804 2570.411 || 412.947 -> 5781.264 MByte/s - average, method 2 : 0.040 0.624 9.865 133.473 1343.250 2767.608 || 543.746 -> 7612.451 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 0.720 0.051 0.051 0.051 0.051 0.012 0.040 2 1.411 0.101 0.101 0.101 0.101 0.024 0.074 4 2.917 0.208 0.208 0.209 0.208 0.056 0.153 8 5.952 0.425 0.425 0.426 0.425 0.112 0.309 16 11.911 0.851 0.850 0.851 0.851 0.225 0.624 32 23.769 1.698 1.697 1.699 1.698 0.449 1.242 64 47.354 3.382 3.381 3.384 3.382 0.897 2.478 128 93.869 6.705 6.702 6.707 6.705 1.775 4.911 256 187.638 13.403 13.396 13.409 13.403 3.558 9.865 512 375.311 26.808 26.785 26.831 26.808 7.128 19.975 1024 747.281 53.377 53.372 53.382 53.377 14.253 39.317 2048 1147.203 81.943 81.594 82.294 81.943 28.393 66.707 4096 2288.023 163.430 163.077 163.785 163.430 56.772 133.473 8933 2982.800 213.057 213.101 213.014 213.057 101.190 186.445 19484 7793.573 556.684 556.788 556.580 556.684 207.731 486.095 42495 11640.253 831.447 835.882 827.035 831.447 375.853 757.698 92682 20343.451 1453.104 1457.635 1448.586 1453.104 712.804 1343.250 202141 23547.825 1681.987 1684.080 1679.897 1681.987 1203.541 1499.763 440872 39563.555 2825.968 2827.444 2824.493 2825.968 1708.585 2415.833 961548 29095.195 2078.228 2033.099 2124.359 2061.282 1614.472 1665.841 2097152 49177.903 3512.707 3543.919 3481.771 3501.554 2570.411 2767.608 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-7*2fix : 0.051 0.846 13.330 163.903 1459.576 3533.605 -> 650.153 -> 9102.148 MByte/s p01 ring-3*4&+1 : 0.051 0.852 13.409 164.623 1465.606 3544.762 -> 652.094 -> 9129.309 MByte/s p02 ring-2*7fix : 0.052 0.852 13.411 158.396 1461.790 3537.778 -> 637.351 -> 8922.914 MByte/s p03 ring-1*14fix : 0.052 0.851 13.414 161.375 1454.055 3555.845 -> 635.565 -> 8897.910 MByte/s p04 ring-1*14fix : 0.052 0.851 13.406 165.093 1452.874 3543.491 -> 635.174 -> 8892.436 MByte/s p05 ring-1*14fix : 0.052 0.850 13.409 165.181 1451.964 3548.073 -> 649.291 -> 9090.076 MByte/s p06 random-cyc-1dim : 0.051 0.851 13.396 164.565 1452.365 3533.176 -> 639.075 -> 8947.044 MByte/s p07 random-cyc-1dim : 0.051 0.852 13.404 164.165 1452.191 3568.455 -> 649.846 -> 9097.846 MByte/s p08 random-cyc-1dim : 0.051 0.852 13.407 164.911 1445.774 3557.752 -> 647.592 -> 9066.281 MByte/s p09 random-cyc-1dim : 0.052 0.851 13.410 164.734 1450.328 3539.736 -> 648.295 -> 9076.127 MByte/s p10 random-cyc-1dim : 0.051 0.852 13.416 159.559 1451.323 3487.071 -> 630.962 -> 8833.470 MByte/s p11 random-cyc-1dim : 0.052 0.852 13.415 163.557 1449.672 3531.130 -> 648.360 -> 9077.035 MByte/s p12 random-cyc-1dim : 0.052 0.851 13.408 164.695 1448.103 3542.821 -> 649.891 -> 9098.467 MByte/s p13 random-cyc-1dim : 0.052 0.851 13.408 162.601 1445.729 3040.655 -> 624.118 -> 8737.653 MByte/s p14 random-cyc-1dim : 0.052 0.851 13.416 164.422 1441.143 3514.913 -> 647.855 -> 9069.972 MByte/s p15 random-cyc-1dim : 0.051 0.852 13.412 164.713 1449.272 3536.752 -> 649.780 -> 9096.925 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.051 0.851 13.423 164.511 1448.658 3547.352 -> 648.573 -> 9080.024 MByte/s p17 best bi-section : 0.039 0.649 10.134 116.810 1081.502 3427.839 -> 568.668 -> 7961.346 MByte/s p18 worst bi-section : 0.039 0.649 10.131 116.685 1082.403 3429.139 -> 569.135 -> 7967.886 MByte/s p19 acyclic-1dim-all : 0.048 0.790 12.457 153.285 1357.739 3292.811 -> 599.960 -> 8399.433 MByte/s p20 acyclic-2dim-all : 0.039 0.650 10.205 124.801 1195.393 2776.706 -> 518.218 -> 7255.048 MByte/s p21 acyclic-3dim-all : 0.030 0.502 7.883 94.667 946.058 2068.625 -> 411.265 -> 5757.710 MByte/s p22 cyclic-1dim-all : 0.051 0.850 13.411 164.902 1456.498 3028.168 -> 611.933 -> 8567.068 MByte/s p23 cyclic-2dim-all : 0.051 0.853 13.449 164.714 1461.671 3496.870 -> 642.104 -> 8989.460 MByte/s p24 cyclic-3dim-all : 0.044 0.732 11.545 141.806 1260.703 3044.646 -> 547.356 -> 7662.978 MByte/s log_avg of all rings : 0.051 0.850 13.396 163.077 1457.635 3543.919 || 643.230 -> 9005.216 MByte/s log_avg of all random : 0.051 0.851 13.409 163.785 1448.586 3481.771 || 643.518 -> 9009.251 MByte/s log_avg(ring,random) : 0.051 0.851 13.403 163.430 1453.104 3512.707 || 643.374 -> 9007.233 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 9007.233 MByte/s on 14 processes ( = 643.374 MByte/s * 14 processes) system parameters : 14 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 = 9007.233 MB/s = 643.374 * 14 PEs with 256 MB/PE on SUPER-UX hwwsx4 9.1 Rev1 SX-4