b_eff = 4535.283 MB/s = 647.898 * 7 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 = 7 1-dim-paterns: size = 7 2-dim-paterns: size = 3 * 2 3-dim-paterns: size = 3 * 2 * 1 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 = 71.278 sec sum of max elapsed time per entries above = 70.474 sec difference = 0.804 sec = 1.1% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-3*2&+1 => 1 sendrecv_calls with 7 messages, i.e. 4.2 msgs/used node, all nodes are used p01 ring-2*4&-1 => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p02 ring-1*7fix => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p03 ring-1*7fix => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p04 ring-1*7fix => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p05 ring-1*7fix => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 6 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p18 worst bi-section => 2 sendrecv_calls with 6 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p19 acyclic-1dim-all => 2 sendrecv_calls with 12 messages, i.e. 4.2 msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p21 acyclic-3dim-all => 4 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p22 cyclic-1dim-all => 2 sendrecv_calls with 14 messages, i.e. 4.2 msgs/used node, all nodes are used p23 cyclic-2dim-all => 3 sendrecv_calls with 18 messages, i.e. 4.2 msgs/used node, 1 nodes are UNUSED p24 cyclic-3dim-all => 3 sendrecv_calls with 18 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-3*2&+1 : 648.662 190.300 528.515 -> 648.662 -> 4540.633 MByte/s p01 ring-2*4&-1 : 653.192 349.698 552.844 -> 653.192 -> 4572.346 MByte/s p02 ring-1*7fix : 622.269 536.934 557.013 -> 622.269 -> 4355.880 MByte/s p03 ring-1*7fix : 650.831 535.390 551.584 -> 650.831 -> 4555.817 MByte/s p04 ring-1*7fix : 649.662 535.459 556.287 -> 649.662 -> 4547.637 MByte/s p05 ring-1*7fix : 649.081 536.928 576.286 -> 649.081 -> 4543.566 MByte/s p06 random-cyc-1dim : 611.261 534.287 549.095 -> 611.261 -> 4278.824 MByte/s p07 random-cyc-1dim : 649.662 534.650 529.392 -> 649.662 -> 4547.636 MByte/s p08 random-cyc-1dim : 635.391 535.220 557.522 -> 635.391 -> 4447.738 MByte/s p09 random-cyc-1dim : 643.581 535.317 554.081 -> 643.581 -> 4505.066 MByte/s p10 random-cyc-1dim : 640.944 537.590 543.736 -> 640.944 -> 4486.611 MByte/s p11 random-cyc-1dim : 646.119 537.476 558.853 -> 646.119 -> 4522.833 MByte/s p12 random-cyc-1dim : 651.426 537.408 562.327 -> 651.426 -> 4559.979 MByte/s p13 random-cyc-1dim : 651.214 537.938 553.021 -> 651.214 -> 4558.498 MByte/s p14 random-cyc-1dim : 651.158 537.367 540.120 -> 651.158 -> 4558.109 MByte/s p15 random-cyc-1dim : 651.805 535.320 571.881 -> 651.805 -> 4562.638 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 650.593 535.788 550.741 -> 650.593 -> 4554.153 MByte/s p17 best bi-section : 347.042 287.579 491.072 -> 491.072 -> 3437.507 MByte/s p18 worst bi-section : 347.610 231.459 491.553 -> 491.553 -> 3440.874 MByte/s p19 acyclic-1dim-all : 530.980 465.268 481.646 -> 530.980 -> 3716.859 MByte/s p20 acyclic-2dim-all : 355.943 351.377 415.192 -> 415.192 -> 2906.344 MByte/s p21 acyclic-3dim-all : 356.031 351.558 402.788 -> 402.788 -> 2819.517 MByte/s p22 cyclic-1dim-all : 635.733 537.438 575.240 -> 635.733 -> 4450.129 MByte/s p23 cyclic-2dim-all : 561.804 448.996 481.239 -> 561.804 -> 3932.629 MByte/s p24 cyclic-3dim-all : 561.144 448.937 474.182 -> 561.144 -> 3928.007 MByte/s log_avg of all rings : 645.528 420.140 553.578 || 645.528 -> 4518.699 MByte/s log_avg of all random : 643.144 536.256 551.883 || 643.144 -> 4502.010 MByte/s log_avg(ring,random) : 644.335 474.660 552.730 ||(644.335 -> 4510.347)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-3*2&+1 : 587.443 646.360 611.455 -> 646.360 -> 4524.521 MByte/s p01 ring-2*4&-1 : 650.252 620.855 607.022 -> 650.252 -> 4551.767 MByte/s p02 ring-1*7fix : 616.186 633.482 616.766 -> 633.482 -> 4434.375 MByte/s p03 ring-1*7fix : 648.487 635.596 627.126 -> 648.487 -> 4539.406 MByte/s p04 ring-1*7fix : 640.325 638.638 622.228 -> 640.325 -> 4482.273 MByte/s p05 ring-1*7fix : 631.025 625.830 606.964 -> 631.025 -> 4417.178 MByte/s p06 random-cyc-1dim : 617.608 629.984 595.247 -> 629.984 -> 4409.889 MByte/s p07 random-cyc-1dim : 642.798 629.851 635.317 -> 642.798 -> 4499.587 MByte/s p08 random-cyc-1dim : 642.126 642.554 603.405 -> 642.554 -> 4497.876 MByte/s p09 random-cyc-1dim : 636.732 623.677 606.845 -> 636.732 -> 4457.122 MByte/s p10 random-cyc-1dim : 643.650 646.855 628.935 -> 646.855 -> 4527.982 MByte/s p11 random-cyc-1dim : 634.442 617.031 630.261 -> 634.442 -> 4441.095 MByte/s p12 random-cyc-1dim : 646.091 636.977 623.979 -> 646.091 -> 4522.634 MByte/s p13 random-cyc-1dim : 618.487 633.571 648.288 -> 648.288 -> 4538.015 MByte/s p14 random-cyc-1dim : 622.243 650.820 594.617 -> 650.820 -> 4555.738 MByte/s p15 random-cyc-1dim : 642.674 649.779 610.878 -> 649.779 -> 4548.453 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 644.559 636.604 600.362 -> 644.559 -> 4511.911 MByte/s p17 best bi-section : 491.928 491.385 491.072 -> 491.928 -> 3443.496 MByte/s p18 worst bi-section : 492.617 492.482 455.032 -> 492.617 -> 3448.320 MByte/s p19 acyclic-1dim-all : 550.301 537.715 545.403 -> 550.301 -> 3852.107 MByte/s p20 acyclic-2dim-all : 400.212 418.172 396.649 -> 418.172 -> 2927.205 MByte/s p21 acyclic-3dim-all : 404.652 399.216 396.105 -> 404.652 -> 2832.564 MByte/s p22 cyclic-1dim-all : 643.223 635.093 601.082 -> 643.223 -> 4502.559 MByte/s p23 cyclic-2dim-all : 557.126 560.906 512.579 -> 560.906 -> 3926.341 MByte/s p24 cyclic-3dim-all : 559.392 560.511 516.549 -> 560.511 -> 3923.577 MByte/s log_avg of all rings : 628.567 633.406 615.214 || 641.613 -> 4491.291 MByte/s log_avg of all random : 634.597 636.018 617.537 || 642.800 -> 4499.598 MByte/s log_avg(ring,random) : 631.575 634.711 616.374 ||(642.206 -> 4495.442)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-3*2&+1 p00 method 0 : 0.051 0.847 13.352 164.255 1446.742 3510.794 -> 648.662 -> 4540.633 MByte/s p00 method 1 : 0.009 0.173 2.743 43.379 468.401 884.674 -> 190.300 -> 1332.097 MByte/s p00 method 2 : 0.035 0.572 8.947 126.712 1214.317 2884.793 -> 528.515 -> 3699.605 MByte/s p01 ring-2*4&-1 p01 method 0 : 0.051 0.851 13.418 164.932 1453.856 3547.952 -> 653.192 -> 4572.346 MByte/s p01 method 1 : 0.017 0.343 5.404 86.209 916.077 1465.011 -> 349.698 -> 2447.887 MByte/s p01 method 2 : 0.040 0.666 10.440 139.891 1354.107 2879.833 -> 552.844 -> 3869.911 MByte/s p02 ring-1*7fix p02 method 0 : 0.052 0.852 13.405 157.972 1444.393 3168.535 -> 622.269 -> 4355.880 MByte/s p02 method 1 : 0.017 0.344 5.422 86.149 913.099 3296.413 -> 536.934 -> 3758.535 MByte/s p02 method 2 : 0.040 0.666 10.430 140.576 1361.493 2811.512 -> 557.013 -> 3899.094 MByte/s p03 ring-1*7fix p03 method 0 : 0.051 0.852 13.418 161.781 1447.186 3546.344 -> 650.831 -> 4555.817 MByte/s p03 method 1 : 0.017 0.343 5.396 85.845 915.666 3302.165 -> 535.390 -> 3747.732 MByte/s p03 method 2 : 0.040 0.666 10.433 140.807 1361.733 2629.038 -> 551.584 -> 3861.089 MByte/s p04 ring-1*7fix p04 method 0 : 0.051 0.852 13.404 163.897 1443.624 3535.582 -> 649.662 -> 4547.637 MByte/s p04 method 1 : 0.017 0.344 5.414 85.957 918.859 3310.463 -> 535.459 -> 3748.210 MByte/s p04 method 2 : 0.040 0.665 10.446 140.843 1367.622 2841.515 -> 556.287 -> 3894.011 MByte/s p05 ring-1*7fix p05 method 0 : 0.051 0.851 13.415 164.989 1447.679 3563.507 -> 649.081 -> 4543.566 MByte/s p05 method 1 : 0.017 0.343 5.410 86.187 918.045 3313.978 -> 536.928 -> 3758.497 MByte/s p05 method 2 : 0.040 0.665 10.439 140.913 1366.877 3153.687 -> 576.286 -> 4034.005 MByte/s p06 random-cyc-1dim p06 method 0 : 0.051 0.851 13.408 164.697 1447.305 2880.308 -> 611.261 -> 4278.824 MByte/s p06 method 1 : 0.017 0.337 5.319 84.629 919.602 3294.446 -> 534.287 -> 3740.008 MByte/s p06 method 2 : 0.040 0.667 10.436 140.924 1367.460 2852.197 -> 549.095 -> 3843.665 MByte/s p07 random-cyc-1dim p07 method 0 : 0.052 0.851 13.408 163.008 1448.811 3569.135 -> 649.662 -> 4547.636 MByte/s p07 method 1 : 0.017 0.338 5.346 85.103 913.942 3299.318 -> 534.650 -> 3742.550 MByte/s p07 method 2 : 0.040 0.664 10.481 142.045 1366.083 2590.599 -> 529.392 -> 3705.741 MByte/s p08 random-cyc-1dim p08 method 0 : 0.051 0.851 13.413 162.376 1446.685 3562.368 -> 635.391 -> 4447.738 MByte/s p08 method 1 : 0.017 0.344 5.428 86.260 920.924 3294.632 -> 535.220 -> 3746.539 MByte/s p08 method 2 : 0.040 0.666 10.439 137.906 1367.431 2865.259 -> 557.522 -> 3902.656 MByte/s p09 random-cyc-1dim p09 method 0 : 0.051 0.851 13.417 164.087 1444.500 3565.833 -> 643.581 -> 4505.066 MByte/s p09 method 1 : 0.017 0.343 5.423 86.163 921.520 3285.321 -> 535.317 -> 3747.219 MByte/s p09 method 2 : 0.040 0.667 10.358 138.444 1367.314 2886.874 -> 554.081 -> 3878.569 MByte/s p10 random-cyc-1dim p10 method 0 : 0.051 0.852 13.425 163.668 1444.815 3565.857 -> 640.944 -> 4486.611 MByte/s p10 method 1 : 0.017 0.344 5.432 86.005 914.293 3319.812 -> 537.590 -> 3763.130 MByte/s p10 method 2 : 0.040 0.665 10.286 140.153 1367.365 2663.740 -> 543.736 -> 3806.154 MByte/s p11 random-cyc-1dim p11 method 0 : 0.052 0.851 13.409 161.518 1459.935 3564.766 -> 646.119 -> 4522.833 MByte/s p11 method 1 : 0.017 0.344 5.411 86.407 922.448 3309.335 -> 537.476 -> 3762.329 MByte/s p11 method 2 : 0.040 0.666 10.429 137.594 1369.890 2882.905 -> 558.853 -> 3911.970 MByte/s p12 random-cyc-1dim p12 method 0 : 0.051 0.852 13.422 164.945 1446.820 3561.666 -> 651.426 -> 4559.979 MByte/s p12 method 1 : 0.017 0.343 5.427 86.269 925.781 3311.676 -> 537.408 -> 3761.859 MByte/s p12 method 2 : 0.040 0.666 10.440 141.207 1370.269 2895.259 -> 562.327 -> 3936.289 MByte/s p13 random-cyc-1dim p13 method 0 : 0.051 0.851 13.389 161.337 1452.328 3567.580 -> 651.214 -> 4558.498 MByte/s p13 method 1 : 0.017 0.343 5.423 86.152 920.613 3304.767 -> 537.938 -> 3765.563 MByte/s p13 method 2 : 0.040 0.666 10.443 141.428 1367.090 2602.585 -> 553.021 -> 3871.146 MByte/s p14 random-cyc-1dim p14 method 0 : 0.051 0.851 13.413 160.737 1443.845 3567.240 -> 651.158 -> 4558.109 MByte/s p14 method 1 : 0.017 0.346 5.421 86.118 917.188 3301.083 -> 537.367 -> 3761.569 MByte/s p14 method 2 : 0.040 0.665 10.422 139.972 1365.852 2539.528 -> 540.120 -> 3780.843 MByte/s p15 random-cyc-1dim p15 method 0 : 0.051 0.851 13.421 161.832 1459.125 3562.465 -> 651.805 -> 4562.638 MByte/s p15 method 1 : 0.017 0.341 5.439 86.224 919.652 3285.300 -> 535.320 -> 3747.242 MByte/s p15 method 2 : 0.040 0.665 10.453 140.678 1368.187 2883.603 -> 571.881 -> 4003.170 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.052 0.851 13.435 164.679 1449.070 3553.821 -> 650.593 -> 4554.153 MByte/s p16 method 1 : 0.017 0.343 5.439 86.335 925.603 3301.978 -> 535.788 -> 3750.518 MByte/s p16 method 2 : 0.039 0.651 10.212 138.330 1366.584 2890.359 -> 550.741 -> 3855.187 MByte/s p17 best bi-section p17 method 0 : 0.034 0.557 8.702 100.264 719.725 2092.871 -> 347.042 -> 2429.294 MByte/s p17 method 1 : 0.008 0.148 2.365 37.690 508.406 1418.259 -> 287.579 -> 2013.051 MByte/s p17 method 2 : 0.023 0.379 5.957 85.621 939.556 2981.461 -> 491.072 -> 3437.507 MByte/s p18 worst bi-section p18 method 0 : 0.034 0.556 8.701 100.254 720.540 2090.613 -> 347.610 -> 2433.268 MByte/s p18 method 1 : 0.008 0.149 2.366 37.610 507.884 679.582 -> 231.459 -> 1620.214 MByte/s p18 method 2 : 0.023 0.379 5.954 85.695 938.817 2982.886 -> 491.553 -> 3440.874 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.044 0.731 11.511 138.999 1254.038 3068.030 -> 530.980 -> 3716.859 MByte/s p19 method 1 : 0.015 0.293 4.646 73.763 806.856 2842.134 -> 465.268 -> 3256.877 MByte/s p19 method 2 : 0.031 0.561 8.923 120.718 1172.298 2438.691 -> 481.646 -> 3371.523 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.029 0.480 7.533 92.217 771.979 2066.388 -> 355.943 -> 2491.600 MByte/s p20 method 1 : 0.017 0.356 5.557 88.443 795.335 1766.987 -> 351.377 -> 2459.641 MByte/s p20 method 2 : 0.023 0.383 5.949 82.710 886.688 2340.687 -> 415.192 -> 2906.344 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.029 0.480 7.461 91.368 771.871 2067.268 -> 356.031 -> 2492.215 MByte/s p21 method 1 : 0.017 0.355 5.560 87.893 796.293 1775.995 -> 351.558 -> 2460.906 MByte/s p21 method 2 : 0.023 0.379 5.855 82.049 883.878 2102.518 -> 402.788 -> 2819.517 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.051 0.851 13.409 162.010 1458.327 3551.894 -> 635.733 -> 4450.129 MByte/s p22 method 1 : 0.017 0.345 5.414 86.072 924.596 3325.920 -> 537.438 -> 3762.069 MByte/s p22 method 2 : 0.040 0.665 10.403 140.723 1366.852 2939.361 -> 575.240 -> 4026.679 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.044 0.731 11.539 141.791 1257.023 3065.735 -> 561.804 -> 3932.629 MByte/s p23 method 1 : 0.022 0.461 7.214 114.284 992.029 2280.603 -> 448.996 -> 3142.972 MByte/s p23 method 2 : 0.033 0.545 8.539 118.136 1248.579 2352.563 -> 481.239 -> 3368.670 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.044 0.731 11.519 141.555 1256.185 3062.392 -> 561.144 -> 3928.007 MByte/s p24 method 1 : 0.022 0.461 7.201 113.791 993.407 2284.134 -> 448.937 -> 3142.561 MByte/s p24 method 2 : 0.033 0.544 8.532 118.062 1255.910 2451.850 -> 474.182 -> 3319.272 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.851 13.402 162.952 1447.243 3475.843 || 645.528 -> 4518.699 MByte/s - ring, method 1 : 0.015 0.306 4.830 76.781 819.382 2317.131 || 420.140 -> 2940.980 MByte/s - ring, method 2 : 0.039 0.649 10.173 138.188 1336.487 2862.626 || 553.578 -> 3875.049 MByte/s log_avg of all random - random, method 0 : 0.051 0.851 13.413 162.814 1449.406 3489.962 || 643.144 -> 4502.010 MByte/s - random, method 1 : 0.017 0.342 5.407 85.931 919.590 3300.552 || 536.256 -> 3753.789 MByte/s - random, method 2 : 0.040 0.666 10.418 140.027 1367.693 2762.660 || 551.883 -> 3863.184 MByte/s log_avg(ring,random) - average, method 0 : 0.051 0.851 13.407 162.883 1448.324 3482.895 || 644.335 -> 4510.347 MByte/s - average, method 1 : 0.016 0.324 5.110 81.227 868.041 2765.467 || 474.660 -> 3322.622 MByte/s - average, method 2 : 0.039 0.657 10.295 139.105 1352.000 2812.199 || 552.730 -> 3869.112 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 0.359 0.051 0.051 0.051 0.051 0.016 0.039 2 0.711 0.102 0.102 0.102 0.102 0.032 0.078 4 1.458 0.208 0.208 0.208 0.208 0.081 0.162 8 2.976 0.425 0.425 0.425 0.425 0.162 0.330 16 5.957 0.851 0.851 0.851 0.851 0.324 0.657 32 11.869 1.696 1.695 1.696 1.696 0.647 1.315 64 23.658 3.380 3.380 3.380 3.380 1.287 2.607 128 46.828 6.690 6.674 6.705 6.690 2.558 5.137 256 93.851 13.407 13.402 13.413 13.407 5.110 10.295 512 187.697 26.814 26.795 26.832 26.814 10.231 20.553 1024 373.780 53.397 53.363 53.431 53.397 20.445 41.028 2048 574.696 82.099 82.123 82.075 82.099 40.764 69.661 4096 1140.183 162.883 162.952 162.814 162.883 81.227 139.105 8933 1478.795 211.256 211.017 211.496 211.256 135.232 187.559 19484 3888.511 555.502 557.601 553.410 555.502 275.658 504.905 42495 5803.764 829.109 831.522 826.703 829.109 478.742 762.050 92682 10138.270 1448.324 1447.243 1449.406 1448.324 868.041 1352.000 202141 11829.532 1689.933 1690.237 1689.629 1689.933 1376.831 1527.590 440872 19802.074 2828.868 2830.043 2827.693 2828.868 2035.125 2421.988 961548 15202.861 2171.837 2185.353 2158.405 2128.096 1799.562 1741.707 2097152 24625.641 3517.949 3498.840 3537.162 3482.895 2765.467 2812.199 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-3*2&+1 : 0.051 0.847 13.352 164.255 1446.742 3510.794 -> 648.662 -> 4540.633 MByte/s p01 ring-2*4&-1 : 0.051 0.851 13.418 164.932 1453.856 3547.952 -> 653.192 -> 4572.346 MByte/s p02 ring-1*7fix : 0.052 0.852 13.405 157.972 1444.393 3296.413 -> 635.829 -> 4450.801 MByte/s p03 ring-1*7fix : 0.051 0.852 13.418 161.781 1447.186 3546.344 -> 650.831 -> 4555.817 MByte/s p04 ring-1*7fix : 0.051 0.852 13.404 163.897 1443.624 3535.582 -> 649.662 -> 4547.637 MByte/s p05 ring-1*7fix : 0.051 0.851 13.415 164.989 1447.679 3563.507 -> 649.081 -> 4543.566 MByte/s p06 random-cyc-1dim : 0.051 0.851 13.408 164.697 1447.305 3294.446 -> 633.798 -> 4436.585 MByte/s p07 random-cyc-1dim : 0.052 0.851 13.408 163.008 1448.811 3569.135 -> 649.662 -> 4547.636 MByte/s p08 random-cyc-1dim : 0.051 0.851 13.413 162.376 1446.685 3562.368 -> 645.222 -> 4516.553 MByte/s p09 random-cyc-1dim : 0.051 0.851 13.417 164.087 1444.500 3565.833 -> 646.565 -> 4525.952 MByte/s p10 random-cyc-1dim : 0.051 0.852 13.425 163.668 1444.815 3565.857 -> 648.551 -> 4539.857 MByte/s p11 random-cyc-1dim : 0.052 0.851 13.409 161.518 1459.935 3564.766 -> 650.241 -> 4551.685 MByte/s p12 random-cyc-1dim : 0.051 0.852 13.422 164.945 1446.820 3561.666 -> 651.426 -> 4559.979 MByte/s p13 random-cyc-1dim : 0.051 0.851 13.389 161.337 1452.328 3567.580 -> 651.214 -> 4558.498 MByte/s p14 random-cyc-1dim : 0.051 0.851 13.413 160.737 1443.845 3567.240 -> 651.158 -> 4558.109 MByte/s p15 random-cyc-1dim : 0.051 0.851 13.421 161.832 1459.125 3562.465 -> 651.805 -> 4562.638 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.052 0.851 13.435 164.679 1449.070 3553.821 -> 650.593 -> 4554.153 MByte/s p17 best bi-section : 0.034 0.557 8.702 100.264 939.556 2981.461 -> 493.200 -> 3452.403 MByte/s p18 worst bi-section : 0.034 0.556 8.701 100.254 938.817 2982.886 -> 493.677 -> 3455.740 MByte/s p19 acyclic-1dim-all : 0.044 0.731 11.511 138.999 1254.038 3068.030 -> 551.545 -> 3860.817 MByte/s p20 acyclic-2dim-all : 0.029 0.480 7.533 92.217 886.688 2340.687 -> 418.209 -> 2927.460 MByte/s p21 acyclic-3dim-all : 0.029 0.480 7.461 91.368 883.878 2102.518 -> 405.661 -> 2839.629 MByte/s p22 cyclic-1dim-all : 0.051 0.851 13.409 162.010 1458.327 3551.894 -> 647.533 -> 4532.730 MByte/s p23 cyclic-2dim-all : 0.044 0.731 11.539 141.791 1257.023 3065.735 -> 561.909 -> 3933.365 MByte/s p24 cyclic-3dim-all : 0.044 0.731 11.519 141.555 1256.185 3062.392 -> 561.315 -> 3929.206 MByte/s log_avg of all rings : 0.051 0.851 13.402 162.952 1447.243 3498.840 || 647.852 -> 4534.963 MByte/s log_avg of all random : 0.051 0.851 13.413 162.814 1449.406 3537.162 || 647.943 -> 4535.604 MByte/s log_avg(ring,random) : 0.051 0.851 13.407 162.883 1448.324 3517.949 || 647.898 -> 4535.283 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 4535.283 MByte/s on 7 processes ( = 647.898 MByte/s * 7 processes) system parameters : 7 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 = 4535.283 MB/s = 647.898 * 7 PEs with 256 MB/PE on SUPER-UX hwwsx4 9.1 Rev1 SX-4