b_eff = 7738.770 MB/s = 644.898 * 12 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 = 12 1-dim-paterns: size = 12 2-dim-paterns: size = 4 * 3 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 = 87.015 sec sum of max elapsed time per entries above = 86.844 sec difference = 0.171 sec = 0.2% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-6*2fix => 1 sendrecv_calls with 12 messages, i.e. 4.2 msgs/used node, all nodes are used p01 ring-3*4fix => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p02 ring-1*12fix => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p03 ring-1*12fix => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p04 ring-1*12fix => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p05 ring-1*12fix => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 12 messages, i.e. 4.2 msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 12 messages, i.e. 4.2 msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 22 messages, i.e. 4.2 msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 34 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, all nodes are used p22 cyclic-1dim-all => 2 sendrecv_calls with 24 messages, i.e. 4.2 msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 48 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, all nodes are used 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-6*2fix : 647.704 292.448 599.780 -> 647.704 -> 7772.451 MByte/s p01 ring-3*4fix : 651.047 336.609 553.648 -> 651.047 -> 7812.561 MByte/s p02 ring-1*12fix : 647.933 474.483 549.225 -> 647.933 -> 7775.193 MByte/s p03 ring-1*12fix : 645.672 471.376 543.749 -> 645.672 -> 7748.065 MByte/s p04 ring-1*12fix : 650.955 475.028 542.333 -> 650.955 -> 7811.457 MByte/s p05 ring-1*12fix : 617.142 475.312 545.489 -> 617.142 -> 7405.707 MByte/s p06 random-cyc-1dim : 646.659 476.159 566.560 -> 646.659 -> 7759.904 MByte/s p07 random-cyc-1dim : 639.092 475.478 533.961 -> 639.092 -> 7669.106 MByte/s p08 random-cyc-1dim : 647.030 474.029 538.089 -> 647.030 -> 7764.360 MByte/s p09 random-cyc-1dim : 648.881 472.591 536.837 -> 648.881 -> 7786.572 MByte/s p10 random-cyc-1dim : 647.048 474.427 527.374 -> 647.048 -> 7764.575 MByte/s p11 random-cyc-1dim : 647.260 472.304 535.704 -> 647.260 -> 7767.123 MByte/s p12 random-cyc-1dim : 628.439 472.973 525.021 -> 628.439 -> 7541.274 MByte/s p13 random-cyc-1dim : 643.269 473.466 538.482 -> 643.269 -> 7719.228 MByte/s p14 random-cyc-1dim : 634.505 474.464 525.987 -> 634.505 -> 7614.062 MByte/s p15 random-cyc-1dim : 646.484 472.417 541.894 -> 646.484 -> 7757.804 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 647.370 473.861 542.421 -> 647.370 -> 7768.438 MByte/s p17 best bi-section : 404.807 291.776 566.989 -> 566.989 -> 6803.867 MByte/s p18 worst bi-section : 404.696 218.320 566.287 -> 566.287 -> 6795.447 MByte/s p19 acyclic-1dim-all : 583.193 438.141 525.895 -> 583.193 -> 6998.320 MByte/s p20 acyclic-2dim-all : 437.855 427.184 401.082 -> 437.855 -> 5254.256 MByte/s p21 acyclic-3dim-all : 404.937 394.265 480.577 -> 480.577 -> 5766.925 MByte/s p22 cyclic-1dim-all : 649.273 473.759 576.056 -> 649.273 -> 7791.272 MByte/s p23 cyclic-2dim-all : 583.485 470.776 510.503 -> 583.485 -> 7001.821 MByte/s p24 cyclic-3dim-all : 606.606 469.017 491.800 -> 606.606 -> 7279.272 MByte/s log_avg of all rings : 643.296 413.120 555.355 || 643.296 -> 7719.557 MByte/s log_avg of all random : 642.835 473.829 536.875 || 642.835 -> 7714.015 MByte/s log_avg(ring,random) : 643.065 442.435 546.037 ||(643.065 -> 7716.786)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-6*2fix : 641.476 643.839 643.461 -> 643.839 -> 7726.073 MByte/s p01 ring-3*4fix : 615.748 615.102 627.678 -> 627.678 -> 7532.132 MByte/s p02 ring-1*12fix : 604.860 632.082 631.785 -> 632.082 -> 7584.983 MByte/s p03 ring-1*12fix : 595.079 633.162 585.549 -> 633.162 -> 7597.947 MByte/s p04 ring-1*12fix : 617.394 649.479 601.585 -> 649.479 -> 7793.753 MByte/s p05 ring-1*12fix : 607.045 625.421 612.584 -> 625.421 -> 7505.058 MByte/s p06 random-cyc-1dim : 616.029 617.638 605.000 -> 617.638 -> 7411.661 MByte/s p07 random-cyc-1dim : 583.694 602.218 630.753 -> 630.753 -> 7569.039 MByte/s p08 random-cyc-1dim : 621.576 595.864 601.888 -> 621.576 -> 7458.910 MByte/s p09 random-cyc-1dim : 605.443 642.318 608.107 -> 642.318 -> 7707.813 MByte/s p10 random-cyc-1dim : 613.266 614.374 617.473 -> 617.473 -> 7409.670 MByte/s p11 random-cyc-1dim : 599.645 629.769 583.888 -> 629.769 -> 7557.232 MByte/s p12 random-cyc-1dim : 611.801 624.044 612.925 -> 624.044 -> 7488.524 MByte/s p13 random-cyc-1dim : 603.156 581.746 600.281 -> 603.156 -> 7237.869 MByte/s p14 random-cyc-1dim : 632.002 594.862 601.978 -> 632.002 -> 7584.027 MByte/s p15 random-cyc-1dim : 625.122 619.992 583.909 -> 625.122 -> 7501.466 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 590.765 623.708 608.411 -> 623.708 -> 7484.499 MByte/s p17 best bi-section : 507.781 547.370 537.512 -> 547.370 -> 6568.439 MByte/s p18 worst bi-section : 551.916 561.503 557.839 -> 561.503 -> 6738.039 MByte/s p19 acyclic-1dim-all : 555.272 590.873 590.071 -> 590.873 -> 7090.474 MByte/s p20 acyclic-2dim-all : 448.043 451.436 455.573 -> 455.573 -> 5466.872 MByte/s p21 acyclic-3dim-all : 467.011 478.518 448.293 -> 478.518 -> 5742.217 MByte/s p22 cyclic-1dim-all : 589.890 621.272 637.116 -> 637.116 -> 7645.396 MByte/s p23 cyclic-2dim-all : 563.653 572.242 543.711 -> 572.242 -> 6866.910 MByte/s p24 cyclic-3dim-all : 570.242 580.895 566.476 -> 580.895 -> 6970.735 MByte/s log_avg of all rings : 613.432 633.080 616.797 || 635.219 -> 7622.628 MByte/s log_avg of all random : 611.029 612.033 604.470 || 624.305 -> 7491.655 MByte/s log_avg(ring,random) : 612.229 622.468 610.602 ||(629.738 -> 7556.857)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-6*2fix p00 method 0 : 0.051 0.846 13.342 164.357 1454.628 3507.412 -> 647.704 -> 7772.451 MByte/s p00 method 1 : 0.007 0.128 2.005 31.360 477.910 1548.061 -> 292.448 -> 3509.376 MByte/s p00 method 2 : 0.035 0.556 8.631 120.929 1220.446 3450.762 -> 599.780 -> 7197.356 MByte/s p01 ring-3*4fix p01 method 0 : 0.051 0.851 13.412 164.970 1456.910 3533.748 -> 651.047 -> 7812.561 MByte/s p01 method 1 : 0.013 0.254 3.959 64.033 774.145 1690.385 -> 336.609 -> 4039.302 MByte/s p01 method 2 : 0.039 0.660 10.230 136.985 1375.646 2607.581 -> 553.648 -> 6643.779 MByte/s p02 ring-1*12fix p02 method 0 : 0.052 0.849 13.419 157.461 1440.508 3544.810 -> 647.933 -> 7775.193 MByte/s p02 method 1 : 0.013 0.245 4.028 63.644 778.372 3108.863 -> 474.483 -> 5693.792 MByte/s p02 method 2 : 0.038 0.651 10.145 124.763 1367.985 2654.771 -> 549.225 -> 6590.705 MByte/s p03 ring-1*12fix p03 method 0 : 0.052 0.850 13.421 156.006 1448.197 3571.201 -> 645.672 -> 7748.065 MByte/s p03 method 1 : 0.013 0.254 4.057 64.109 768.767 3045.743 -> 471.376 -> 5656.518 MByte/s p03 method 2 : 0.039 0.651 10.057 135.407 1366.570 2603.075 -> 543.749 -> 6524.985 MByte/s p04 ring-1*12fix p04 method 0 : 0.052 0.850 13.432 163.969 1441.587 3572.515 -> 650.955 -> 7811.457 MByte/s p04 method 1 : 0.013 0.256 3.894 62.821 772.073 3125.785 -> 475.028 -> 5700.331 MByte/s p04 method 2 : 0.040 0.658 10.152 135.099 1366.123 2636.720 -> 542.333 -> 6507.992 MByte/s p05 ring-1*12fix p05 method 0 : 0.052 0.836 13.430 163.996 1442.321 3582.181 -> 617.142 -> 7405.707 MByte/s p05 method 1 : 0.013 0.248 4.031 60.556 767.182 3120.131 -> 475.312 -> 5703.749 MByte/s p05 method 2 : 0.040 0.651 10.183 136.343 1367.061 2631.057 -> 545.489 -> 6545.865 MByte/s p06 random-cyc-1dim p06 method 0 : 0.051 0.850 13.421 161.084 1436.048 3543.181 -> 646.659 -> 7759.904 MByte/s p06 method 1 : 0.013 0.254 4.040 64.069 773.039 3126.885 -> 476.159 -> 5713.913 MByte/s p06 method 2 : 0.039 0.654 10.269 135.692 1363.408 3151.186 -> 566.560 -> 6798.721 MByte/s p07 random-cyc-1dim p07 method 0 : 0.049 0.819 12.698 158.961 1449.309 3551.581 -> 639.092 -> 7669.106 MByte/s p07 method 1 : 0.013 0.255 4.037 61.102 774.537 3102.700 -> 475.478 -> 5705.741 MByte/s p07 method 2 : 0.039 0.654 10.182 135.491 1368.624 2645.114 -> 533.961 -> 6407.531 MByte/s p08 random-cyc-1dim p08 method 0 : 0.049 0.805 12.906 156.795 1439.617 3549.466 -> 647.030 -> 7764.360 MByte/s p08 method 1 : 0.013 0.254 4.033 62.167 772.988 3118.145 -> 474.029 -> 5688.352 MByte/s p08 method 2 : 0.039 0.654 10.045 135.278 1370.785 2774.539 -> 538.089 -> 6457.066 MByte/s p09 random-cyc-1dim p09 method 0 : 0.047 0.852 13.419 162.041 1437.461 3575.439 -> 648.881 -> 7786.572 MByte/s p09 method 1 : 0.013 0.254 4.030 64.135 773.758 3117.051 -> 472.591 -> 5671.097 MByte/s p09 method 2 : 0.039 0.653 10.250 130.890 1368.106 2840.300 -> 536.837 -> 6442.041 MByte/s p10 random-cyc-1dim p10 method 0 : 0.052 0.850 13.179 159.755 1443.322 3519.609 -> 647.048 -> 7764.575 MByte/s p10 method 1 : 0.013 0.255 4.035 64.158 775.847 3113.609 -> 474.427 -> 5693.123 MByte/s p10 method 2 : 0.040 0.648 10.190 122.098 1370.059 2653.494 -> 527.374 -> 6328.482 MByte/s p11 random-cyc-1dim p11 method 0 : 0.051 0.835 13.426 164.898 1453.425 3491.204 -> 647.260 -> 7767.123 MByte/s p11 method 1 : 0.013 0.254 4.040 62.831 776.267 3066.155 -> 472.304 -> 5667.649 MByte/s p11 method 2 : 0.040 0.652 9.995 134.735 1365.856 2710.633 -> 535.704 -> 6428.452 MByte/s p12 random-cyc-1dim p12 method 0 : 0.051 0.850 13.388 162.005 1448.477 3485.216 -> 628.439 -> 7541.274 MByte/s p12 method 1 : 0.013 0.254 4.058 63.369 775.809 3092.251 -> 472.973 -> 5675.677 MByte/s p12 method 2 : 0.039 0.652 10.268 132.857 1369.610 2380.455 -> 525.021 -> 6300.253 MByte/s p13 random-cyc-1dim p13 method 0 : 0.050 0.821 12.709 160.598 1441.041 3520.318 -> 643.269 -> 7719.228 MByte/s p13 method 1 : 0.013 0.249 4.037 61.612 771.685 3118.553 -> 473.466 -> 5681.587 MByte/s p13 method 2 : 0.040 0.654 10.139 136.627 1368.628 2871.316 -> 538.482 -> 6461.789 MByte/s p14 random-cyc-1dim p14 method 0 : 0.050 0.786 12.941 154.074 1430.029 3526.095 -> 634.505 -> 7614.062 MByte/s p14 method 1 : 0.013 0.254 4.032 64.247 774.983 3095.756 -> 474.464 -> 5693.567 MByte/s p14 method 2 : 0.040 0.652 10.023 132.618 1364.098 2503.046 -> 525.987 -> 6311.847 MByte/s p15 random-cyc-1dim p15 method 0 : 0.050 0.851 13.185 162.471 1445.671 3500.715 -> 646.484 -> 7757.804 MByte/s p15 method 1 : 0.013 0.254 4.035 64.170 773.129 3067.895 -> 472.417 -> 5669.010 MByte/s p15 method 2 : 0.039 0.652 10.151 135.983 1372.320 2613.065 -> 541.894 -> 6502.727 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.049 0.851 13.427 164.838 1448.033 3515.903 -> 647.370 -> 7768.438 MByte/s p16 method 1 : 0.013 0.255 4.027 64.283 778.694 3101.965 -> 473.861 -> 5686.329 MByte/s p16 method 2 : 0.039 0.649 10.193 135.751 1369.960 2626.772 -> 542.421 -> 6509.047 MByte/s p17 best bi-section p17 method 0 : 0.039 0.651 10.132 115.210 835.893 2433.659 -> 404.807 -> 4857.687 MByte/s p17 method 1 : 0.007 0.128 2.032 32.415 476.898 1562.669 -> 291.776 -> 3501.316 MByte/s p17 method 2 : 0.027 0.436 6.726 97.786 1092.423 3413.422 -> 566.989 -> 6803.867 MByte/s p18 worst bi-section p18 method 0 : 0.039 0.649 10.124 116.815 837.808 2432.056 -> 404.696 -> 4856.351 MByte/s p18 method 1 : 0.007 0.128 2.030 32.423 477.059 591.445 -> 218.320 -> 2619.838 MByte/s p18 method 2 : 0.025 0.430 6.764 97.730 1090.227 3447.176 -> 566.287 -> 6795.447 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.047 0.780 12.311 151.402 1334.159 3274.538 -> 583.193 -> 6998.320 MByte/s p19 method 1 : 0.012 0.235 3.500 58.811 719.153 2871.466 -> 438.141 -> 5257.698 MByte/s p19 method 2 : 0.034 0.599 9.429 123.690 1254.825 2592.205 -> 525.895 -> 6310.740 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.036 0.600 9.463 115.771 1043.558 2297.389 -> 437.855 -> 5254.256 MByte/s p20 method 1 : 0.018 0.355 5.610 86.764 838.368 2414.139 -> 427.184 -> 5126.203 MByte/s p20 method 2 : 0.028 0.471 7.476 102.224 976.562 1823.873 -> 401.082 -> 4812.981 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.035 0.583 9.156 108.822 884.206 2403.353 -> 404.937 -> 4859.249 MByte/s p21 method 1 : 0.020 0.414 6.476 102.758 965.136 1963.127 -> 394.265 -> 4731.179 MByte/s p21 method 2 : 0.029 0.485 7.475 104.277 1100.375 2545.416 -> 480.577 -> 5766.925 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.052 0.852 13.414 164.644 1453.906 3542.462 -> 649.273 -> 7791.272 MByte/s p22 method 1 : 0.013 0.250 4.039 64.194 774.035 3099.875 -> 473.759 -> 5685.113 MByte/s p22 method 2 : 0.039 0.647 10.154 136.898 1362.249 3120.353 -> 576.056 -> 6912.672 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.051 0.846 13.190 164.101 1448.237 2645.581 -> 583.485 -> 7001.821 MByte/s p23 method 1 : 0.025 0.499 7.824 122.154 1103.240 2345.435 -> 470.776 -> 5649.306 MByte/s p23 method 2 : 0.036 0.584 9.385 121.202 1353.595 2461.122 -> 510.503 -> 6126.030 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.051 0.846 13.295 162.751 1460.939 2663.747 -> 606.606 -> 7279.272 MByte/s p24 method 1 : 0.025 0.499 7.864 122.792 1106.229 2346.374 -> 469.017 -> 5628.203 MByte/s p24 method 2 : 0.035 0.576 8.944 120.405 1310.308 2290.082 -> 491.800 -> 5901.602 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.847 13.410 161.752 1447.344 3551.882 || 643.296 -> 7719.557 MByte/s - ring, method 1 : 0.012 0.225 3.560 56.099 712.773 2495.744 || 413.120 -> 4957.445 MByte/s - ring, method 2 : 0.039 0.637 9.882 131.432 1342.773 2748.802 || 555.355 -> 6664.265 MByte/s log_avg of all random - random, method 0 : 0.050 0.832 13.124 160.241 1442.425 3526.176 || 642.835 -> 7714.015 MByte/s - random, method 1 : 0.013 0.254 4.038 63.176 774.203 3101.834 || 473.829 -> 5685.952 MByte/s - random, method 2 : 0.039 0.653 10.151 133.162 1368.147 2706.900 || 536.875 -> 6442.502 MByte/s log_avg(ring,random) - average, method 0 : 0.051 0.839 13.266 160.995 1444.882 3539.005 || 643.065 -> 7716.786 MByte/s - average, method 1 : 0.012 0.239 3.791 59.532 742.853 2782.334 || 442.435 -> 5309.218 MByte/s - average, method 2 : 0.039 0.645 10.015 132.294 1355.401 2727.771 || 546.037 -> 6552.445 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 0.608 0.051 0.051 0.050 0.051 0.012 0.039 2 1.199 0.100 0.101 0.099 0.100 0.025 0.077 4 2.466 0.206 0.208 0.203 0.206 0.060 0.158 8 5.053 0.421 0.425 0.417 0.421 0.119 0.321 16 10.071 0.839 0.847 0.832 0.839 0.239 0.645 32 20.082 1.674 1.694 1.654 1.674 0.479 1.282 64 39.956 3.330 3.361 3.298 3.330 0.949 2.550 128 79.550 6.629 6.693 6.566 6.629 1.885 5.030 256 159.194 13.266 13.410 13.124 13.266 3.791 10.015 512 316.650 26.387 26.516 26.260 26.387 7.602 20.064 1024 630.189 52.516 52.870 52.164 52.516 15.089 40.228 2048 975.793 81.316 81.889 80.747 81.316 30.430 67.422 4096 1931.939 160.995 161.752 160.241 160.995 59.532 132.294 8933 2499.057 208.255 210.009 206.516 208.255 106.558 187.332 19484 6652.643 554.387 557.161 551.627 554.387 219.806 492.816 42495 9923.846 826.987 832.304 821.705 826.987 396.038 766.688 92682 17338.587 1444.882 1447.344 1442.425 1444.882 742.853 1355.401 202141 19913.300 1659.442 1640.787 1678.309 1643.985 1245.788 1514.850 440872 33827.018 2818.918 2826.297 2811.559 2818.918 1867.836 2414.020 961548 25687.524 2140.627 2151.902 2129.411 2113.538 1770.155 1711.665 2097152 42468.066 3539.005 3551.882 3526.176 3539.005 2782.334 2727.771 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-6*2fix : 0.051 0.846 13.342 164.357 1454.628 3507.412 -> 647.704 -> 7772.451 MByte/s p01 ring-3*4fix : 0.051 0.851 13.412 164.970 1456.910 3533.748 -> 651.047 -> 7812.561 MByte/s p02 ring-1*12fix : 0.052 0.849 13.419 157.461 1440.508 3544.810 -> 647.933 -> 7775.193 MByte/s p03 ring-1*12fix : 0.052 0.850 13.421 156.006 1448.197 3571.201 -> 645.672 -> 7748.065 MByte/s p04 ring-1*12fix : 0.052 0.850 13.432 163.969 1441.587 3572.515 -> 650.955 -> 7811.457 MByte/s p05 ring-1*12fix : 0.052 0.836 13.430 163.996 1442.321 3582.181 -> 633.996 -> 7607.958 MByte/s p06 random-cyc-1dim : 0.051 0.850 13.421 161.084 1436.048 3543.181 -> 646.659 -> 7759.904 MByte/s p07 random-cyc-1dim : 0.049 0.819 12.698 158.961 1449.309 3551.581 -> 639.092 -> 7669.106 MByte/s p08 random-cyc-1dim : 0.049 0.805 12.906 156.795 1439.617 3549.466 -> 647.030 -> 7764.360 MByte/s p09 random-cyc-1dim : 0.047 0.852 13.419 162.041 1437.461 3575.439 -> 648.881 -> 7786.572 MByte/s p10 random-cyc-1dim : 0.052 0.850 13.179 159.755 1443.322 3519.609 -> 647.048 -> 7764.575 MByte/s p11 random-cyc-1dim : 0.051 0.835 13.426 164.898 1453.425 3491.204 -> 647.260 -> 7767.123 MByte/s p12 random-cyc-1dim : 0.051 0.850 13.388 162.005 1448.477 3485.216 -> 636.021 -> 7632.250 MByte/s p13 random-cyc-1dim : 0.050 0.821 12.709 160.598 1441.041 3520.318 -> 643.269 -> 7719.228 MByte/s p14 random-cyc-1dim : 0.050 0.786 12.941 154.074 1430.029 3526.095 -> 634.505 -> 7614.062 MByte/s p15 random-cyc-1dim : 0.050 0.851 13.185 162.471 1445.671 3500.715 -> 646.484 -> 7757.804 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.049 0.851 13.427 164.838 1448.033 3515.903 -> 647.370 -> 7768.438 MByte/s p17 best bi-section : 0.039 0.651 10.132 115.210 1092.423 3413.422 -> 569.638 -> 6835.659 MByte/s p18 worst bi-section : 0.039 0.649 10.124 116.815 1090.227 3447.176 -> 568.918 -> 6827.012 MByte/s p19 acyclic-1dim-all : 0.047 0.780 12.311 151.402 1334.159 3274.538 -> 594.688 -> 7136.256 MByte/s p20 acyclic-2dim-all : 0.036 0.600 9.463 115.771 1043.558 2414.139 -> 458.477 -> 5501.726 MByte/s p21 acyclic-3dim-all : 0.035 0.583 9.156 108.822 1100.375 2545.416 -> 481.512 -> 5778.145 MByte/s p22 cyclic-1dim-all : 0.052 0.852 13.414 164.644 1453.906 3542.462 -> 649.273 -> 7791.272 MByte/s p23 cyclic-2dim-all : 0.051 0.846 13.190 164.101 1448.237 2645.581 -> 583.485 -> 7001.821 MByte/s p24 cyclic-3dim-all : 0.051 0.846 13.295 162.751 1460.939 2663.747 -> 606.606 -> 7279.272 MByte/s log_avg of all rings : 0.051 0.847 13.410 161.752 1447.344 3551.882 || 646.192 -> 7754.301 MByte/s log_avg of all random : 0.050 0.832 13.124 160.241 1442.425 3526.176 || 643.606 -> 7723.271 MByte/s log_avg(ring,random) : 0.051 0.839 13.266 160.995 1444.882 3539.005 || 644.898 -> 7738.770 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 7738.770 MByte/s on 12 processes ( = 644.898 MByte/s * 12 processes) system parameters : 12 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 = 7738.770 MB/s = 644.898 * 12 PEs with 256 MB/PE on SUPER-UX hwwsx4 9.1 Rev1 SX-4