b_eff = 435.041 MB/s = 62.149 * 7 PEs with 1024 MB/PE on HP-UX hwwhpv B.11.00 A 9000/800 SECTION-HEAD-BEGIN b_eff.c, Revision 3.2 from Nov. 07, 1999 MEMORY_PER_PROCESSOR = 1024 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), 10624 (300), 27554 (300), 71468 (117), 185364 (45), 480774 (17), 1246974 (6), 3234251 (2), 8388608 (1), SECTION-HEAD-END SECTION-LOOP-BEGIN measurment loop: elapsed time = 369.249 sec sum of max elapsed time per entries above = 372.744 sec difference = -3.495 sec = 0.9% 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. msgs/used node, all nodes are used p01 ring-2*4&-1 => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p02 ring-1*7fix => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p03 ring-1*7fix => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p04 ring-1*7fix => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p05 ring-1*7fix => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 6 messages, i.e. msgs/used node, 1 nodes are UNUSED p18 worst bi-section => 2 sendrecv_calls with 6 messages, i.e. msgs/used node, 1 nodes are UNUSED p19 acyclic-1dim-all => 2 sendrecv_calls with 12 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 14 messages, i.e. msgs/used node, 1 nodes are UNUSED p21 acyclic-3dim-all => 4 sendrecv_calls with 14 messages, i.e. msgs/used node, 1 nodes are UNUSED p22 cyclic-1dim-all => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 3 sendrecv_calls with 18 messages, i.e. msgs/used node, 1 nodes are UNUSED p24 cyclic-3dim-all => 3 sendrecv_calls with 18 messages, i.e. 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 : 64.630 36.048 46.587 -> 64.630 -> 452.407 MByte/s p01 ring-2*4&-1 : 62.092 56.941 46.538 -> 62.092 -> 434.647 MByte/s p02 ring-1*7fix : 61.802 57.262 47.875 -> 61.802 -> 432.613 MByte/s p03 ring-1*7fix : 61.777 57.362 47.540 -> 61.777 -> 432.438 MByte/s p04 ring-1*7fix : 61.954 57.644 47.258 -> 61.954 -> 433.676 MByte/s p05 ring-1*7fix : 62.042 57.456 47.609 -> 62.042 -> 434.296 MByte/s p06 random-cyc-1dim : 61.797 57.369 46.447 -> 61.797 -> 432.581 MByte/s p07 random-cyc-1dim : 61.760 57.149 47.558 -> 61.760 -> 432.319 MByte/s p08 random-cyc-1dim : 61.983 57.367 47.176 -> 61.983 -> 433.880 MByte/s p09 random-cyc-1dim : 61.951 57.019 45.869 -> 61.951 -> 433.660 MByte/s p10 random-cyc-1dim : 61.617 57.376 45.756 -> 61.617 -> 431.319 MByte/s p11 random-cyc-1dim : 61.705 56.982 45.205 -> 61.705 -> 431.936 MByte/s p12 random-cyc-1dim : 61.904 57.393 46.915 -> 61.904 -> 433.328 MByte/s p13 random-cyc-1dim : 61.674 57.119 47.378 -> 61.674 -> 431.717 MByte/s p14 random-cyc-1dim : 61.495 56.969 47.282 -> 61.495 -> 430.464 MByte/s p15 random-cyc-1dim : 61.614 57.136 46.130 -> 61.614 -> 431.300 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 61.915 57.290 46.451 -> 61.915 -> 433.404 MByte/s p17 best bi-section : 53.040 48.767 41.986 -> 53.040 -> 371.278 MByte/s p18 worst bi-section : 52.482 47.974 42.138 -> 52.482 -> 367.376 MByte/s p19 acyclic-1dim-all : 55.609 51.382 41.307 -> 55.609 -> 389.266 MByte/s p20 acyclic-2dim-all : 44.226 43.142 41.144 -> 44.226 -> 309.581 MByte/s p21 acyclic-3dim-all : 43.908 43.173 41.151 -> 43.908 -> 307.356 MByte/s p22 cyclic-1dim-all : 59.285 57.220 46.098 -> 59.285 -> 414.993 MByte/s p23 cyclic-2dim-all : 54.429 52.985 45.314 -> 54.429 -> 381.002 MByte/s p24 cyclic-3dim-all : 54.563 52.900 45.086 -> 54.563 -> 381.938 MByte/s log_avg of all rings : 62.375 53.066 47.232 || 62.375 -> 436.623 MByte/s log_avg of all random : 61.750 57.188 46.565 || 61.750 -> 432.249 MByte/s log_avg(ring,random) : 62.062 55.088 46.897 ||( 62.062 -> 434.431)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 : 62.443 61.564 62.820 -> 62.820 -> 439.741 MByte/s p01 ring-2*4&-1 : 60.120 60.261 60.282 -> 60.282 -> 421.977 MByte/s p02 ring-1*7fix : 59.642 56.236 61.526 -> 61.526 -> 430.683 MByte/s p03 ring-1*7fix : 59.871 58.128 61.702 -> 61.702 -> 431.914 MByte/s p04 ring-1*7fix : 59.977 60.485 61.595 -> 61.595 -> 431.165 MByte/s p05 ring-1*7fix : 59.988 60.795 61.906 -> 61.906 -> 433.344 MByte/s p06 random-cyc-1dim : 59.812 55.790 61.056 -> 61.056 -> 427.390 MByte/s p07 random-cyc-1dim : 59.847 61.494 57.738 -> 61.494 -> 430.459 MByte/s p08 random-cyc-1dim : 60.478 61.773 56.849 -> 61.773 -> 432.410 MByte/s p09 random-cyc-1dim : 60.180 61.749 56.263 -> 61.749 -> 432.246 MByte/s p10 random-cyc-1dim : 59.496 61.521 58.504 -> 61.521 -> 430.646 MByte/s p11 random-cyc-1dim : 59.820 61.091 59.776 -> 61.091 -> 427.639 MByte/s p12 random-cyc-1dim : 60.130 61.616 60.562 -> 61.616 -> 431.315 MByte/s p13 random-cyc-1dim : 59.658 60.570 60.953 -> 60.953 -> 426.669 MByte/s p14 random-cyc-1dim : 59.586 61.305 60.844 -> 61.305 -> 429.135 MByte/s p15 random-cyc-1dim : 54.274 61.213 61.394 -> 61.394 -> 429.756 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 60.027 59.378 61.446 -> 61.446 -> 430.123 MByte/s p17 best bi-section : 51.873 53.387 53.715 -> 53.715 -> 376.008 MByte/s p18 worst bi-section : 51.835 53.287 52.676 -> 53.287 -> 373.010 MByte/s p19 acyclic-1dim-all : 49.720 55.342 55.316 -> 55.342 -> 387.393 MByte/s p20 acyclic-2dim-all : 41.545 44.435 41.724 -> 44.435 -> 311.042 MByte/s p21 acyclic-3dim-all : 42.331 42.023 41.977 -> 42.331 -> 296.317 MByte/s p22 cyclic-1dim-all : 55.192 51.655 56.165 -> 56.165 -> 393.158 MByte/s p23 cyclic-2dim-all : 52.942 51.662 53.596 -> 53.596 -> 375.173 MByte/s p24 cyclic-3dim-all : 53.992 54.027 54.585 -> 54.585 -> 382.094 MByte/s log_avg of all rings : 60.333 59.550 61.634 || 61.634 -> 431.439 MByte/s log_avg of all random : 59.302 60.787 59.366 || 61.395 -> 429.762 MByte/s log_avg(ring,random) : 59.815 60.165 60.489 ||( 61.514 -> 430.600)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 185364 8388608 -> average -> accumulated p00 ring-3*2&+1 p00 method 0 : 0.103 1.405 13.961 43.459 155.197 162.306 -> 64.630 -> 452.407 MByte/s p00 method 1 : 0.014 0.216 2.645 18.083 79.827 163.256 -> 36.048 -> 252.335 MByte/s p00 method 2 : 0.056 0.843 9.455 41.518 122.356 95.195 -> 46.587 -> 326.106 MByte/s p01 ring-2*4&-1 p01 method 0 : 0.103 1.476 13.909 43.135 147.127 162.857 -> 62.092 -> 434.647 MByte/s p01 method 1 : 0.027 0.432 5.669 34.588 145.133 161.132 -> 56.941 -> 398.590 MByte/s p01 method 2 : 0.055 0.832 9.408 39.030 115.849 92.031 -> 46.538 -> 325.763 MByte/s p02 ring-1*7fix p02 method 0 : 0.103 1.486 13.823 43.184 146.056 162.246 -> 61.802 -> 432.613 MByte/s p02 method 1 : 0.027 0.428 5.529 34.816 144.806 162.245 -> 57.262 -> 400.831 MByte/s p02 method 2 : 0.055 0.830 9.446 39.195 114.080 119.911 -> 47.875 -> 335.125 MByte/s p03 ring-1*7fix p03 method 0 : 0.104 1.480 13.793 43.030 146.857 161.034 -> 61.777 -> 432.438 MByte/s p03 method 1 : 0.027 0.431 5.676 34.894 145.215 161.245 -> 57.362 -> 401.534 MByte/s p03 method 2 : 0.055 0.832 9.332 39.089 112.889 117.437 -> 47.540 -> 332.783 MByte/s p04 ring-1*7fix p04 method 0 : 0.103 1.483 13.793 43.449 146.525 161.870 -> 61.954 -> 433.676 MByte/s p04 method 1 : 0.027 0.429 5.668 34.799 145.160 162.020 -> 57.644 -> 403.505 MByte/s p04 method 2 : 0.055 0.825 9.340 39.155 112.778 117.908 -> 47.258 -> 330.806 MByte/s p05 ring-1*7fix p05 method 0 : 0.102 1.500 13.722 42.876 147.042 161.845 -> 62.042 -> 434.296 MByte/s p05 method 1 : 0.027 0.433 5.675 34.713 144.608 161.531 -> 57.456 -> 402.189 MByte/s p05 method 2 : 0.055 0.829 9.357 39.077 112.066 116.684 -> 47.609 -> 333.261 MByte/s p06 random-cyc-1dim p06 method 0 : 0.100 1.507 13.746 43.054 147.287 163.486 -> 61.797 -> 432.581 MByte/s p06 method 1 : 0.027 0.424 5.628 34.445 145.089 161.666 -> 57.369 -> 401.584 MByte/s p06 method 2 : 0.055 0.831 9.272 38.672 113.995 92.114 -> 46.447 -> 325.128 MByte/s p07 random-cyc-1dim p07 method 0 : 0.103 1.491 13.738 43.373 147.859 160.752 -> 61.760 -> 432.319 MByte/s p07 method 1 : 0.028 0.432 5.616 35.140 143.622 162.829 -> 57.149 -> 400.044 MByte/s p07 method 2 : 0.055 0.832 9.344 39.045 112.717 114.638 -> 47.558 -> 332.908 MByte/s p08 random-cyc-1dim p08 method 0 : 0.102 1.499 13.554 43.739 147.318 162.213 -> 61.983 -> 433.880 MByte/s p08 method 1 : 0.027 0.432 5.616 34.962 145.357 161.925 -> 57.367 -> 401.568 MByte/s p08 method 2 : 0.055 0.831 9.381 39.299 114.255 119.239 -> 47.176 -> 330.232 MByte/s p09 random-cyc-1dim p09 method 0 : 0.103 1.482 13.808 42.980 147.825 161.332 -> 61.951 -> 433.660 MByte/s p09 method 1 : 0.027 0.431 5.345 35.369 144.086 162.800 -> 57.019 -> 399.132 MByte/s p09 method 2 : 0.055 0.826 9.347 39.199 111.451 91.295 -> 45.869 -> 321.084 MByte/s p10 random-cyc-1dim p10 method 0 : 0.094 1.442 13.627 43.533 145.446 161.027 -> 61.617 -> 431.319 MByte/s p10 method 1 : 0.027 0.429 5.629 35.301 144.706 161.721 -> 57.376 -> 401.630 MByte/s p10 method 2 : 0.053 0.832 9.251 38.936 113.706 91.066 -> 45.756 -> 320.289 MByte/s p11 random-cyc-1dim p11 method 0 : 0.102 1.482 13.723 43.440 147.585 160.170 -> 61.705 -> 431.936 MByte/s p11 method 1 : 0.027 0.423 5.582 34.658 143.271 161.101 -> 56.982 -> 398.872 MByte/s p11 method 2 : 0.055 0.834 9.317 38.892 113.206 90.708 -> 45.205 -> 316.438 MByte/s p12 random-cyc-1dim p12 method 0 : 0.103 1.496 13.820 43.211 146.281 161.185 -> 61.904 -> 433.328 MByte/s p12 method 1 : 0.027 0.429 5.494 35.031 145.268 160.129 -> 57.393 -> 401.754 MByte/s p12 method 2 : 0.055 0.832 9.335 39.515 113.783 116.454 -> 46.915 -> 328.403 MByte/s p13 random-cyc-1dim p13 method 0 : 0.101 1.476 13.829 43.420 145.804 161.363 -> 61.674 -> 431.717 MByte/s p13 method 1 : 0.027 0.434 5.652 35.033 144.113 159.949 -> 57.119 -> 399.830 MByte/s p13 method 2 : 0.055 0.837 9.373 39.373 111.964 114.340 -> 47.378 -> 331.648 MByte/s p14 random-cyc-1dim p14 method 0 : 0.102 1.479 13.772 43.681 145.846 159.391 -> 61.495 -> 430.464 MByte/s p14 method 1 : 0.028 0.429 5.389 35.801 143.963 161.554 -> 56.969 -> 398.783 MByte/s p14 method 2 : 0.055 0.835 9.329 39.263 114.108 114.034 -> 47.282 -> 330.976 MByte/s p15 random-cyc-1dim p15 method 0 : 0.102 1.347 13.788 43.837 146.891 160.925 -> 61.614 -> 431.300 MByte/s p15 method 1 : 0.027 0.429 5.590 34.817 144.960 159.993 -> 57.136 -> 399.954 MByte/s p15 method 2 : 0.055 0.827 9.403 39.212 114.269 90.607 -> 46.130 -> 322.908 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.104 1.487 13.758 43.414 146.907 161.767 -> 61.915 -> 433.404 MByte/s p16 method 1 : 0.028 0.430 5.416 35.236 145.418 161.694 -> 57.290 -> 401.027 MByte/s p16 method 2 : 0.055 0.834 9.376 39.471 116.328 91.375 -> 46.451 -> 325.160 MByte/s p17 best bi-section p17 method 0 : 0.121 1.555 9.116 25.936 134.523 146.253 -> 53.040 -> 371.278 MByte/s p17 method 1 : 0.012 0.189 2.671 23.460 132.231 145.678 -> 48.767 -> 341.372 MByte/s p17 method 2 : 0.032 0.479 5.828 34.484 91.419 149.277 -> 41.986 -> 293.902 MByte/s p18 worst bi-section p18 method 0 : 0.121 1.656 9.148 26.008 132.141 147.001 -> 52.482 -> 367.376 MByte/s p18 method 1 : 0.012 0.189 2.649 23.539 130.363 146.617 -> 47.974 -> 335.821 MByte/s p18 method 2 : 0.032 0.477 5.877 34.541 85.801 147.823 -> 42.138 -> 294.969 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.081 1.129 11.489 39.139 131.375 146.348 -> 55.609 -> 389.266 MByte/s p19 method 1 : 0.024 0.371 4.450 30.273 130.820 146.491 -> 51.382 -> 359.674 MByte/s p19 method 2 : 0.048 0.714 8.038 37.555 100.381 79.488 -> 41.307 -> 289.147 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.066 0.963 7.837 27.220 109.171 117.072 -> 44.226 -> 309.581 MByte/s p20 method 1 : 0.031 0.473 5.030 28.724 106.342 118.663 -> 43.142 -> 301.991 MByte/s p20 method 2 : 0.033 0.498 5.851 30.391 96.581 117.086 -> 41.144 -> 288.009 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.066 0.963 7.889 27.485 108.646 111.427 -> 43.908 -> 307.356 MByte/s p21 method 1 : 0.031 0.477 4.908 28.905 106.545 120.544 -> 43.173 -> 302.208 MByte/s p21 method 2 : 0.033 0.498 5.839 30.320 94.412 117.595 -> 41.151 -> 288.055 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.102 1.457 13.604 44.151 136.213 136.719 -> 59.285 -> 414.993 MByte/s p22 method 1 : 0.027 0.429 5.651 35.153 144.603 158.165 -> 57.220 -> 400.537 MByte/s p22 method 2 : 0.056 0.828 9.309 39.523 113.809 86.702 -> 46.098 -> 322.684 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.088 1.262 12.049 38.481 129.723 141.138 -> 54.429 -> 381.002 MByte/s p23 method 1 : 0.041 0.621 7.275 35.155 131.205 142.658 -> 52.985 -> 370.892 MByte/s p23 method 2 : 0.049 0.733 8.287 37.392 105.769 113.305 -> 45.314 -> 317.196 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.088 1.266 12.009 38.846 130.757 141.404 -> 54.563 -> 381.938 MByte/s p24 method 1 : 0.040 0.627 7.263 34.991 129.473 141.834 -> 52.900 -> 370.301 MByte/s p24 method 2 : 0.048 0.733 8.294 37.400 107.193 111.629 -> 45.086 -> 315.602 MByte/s log_avg of all rings - ring, method 0 : 0.103 1.471 13.833 43.188 148.101 162.026 || 62.375 -> 436.623 MByte/s - ring, method 1 : 0.024 0.384 4.974 31.174 131.258 161.903 || 53.066 -> 371.459 MByte/s - ring, method 2 : 0.055 0.832 9.389 39.501 114.951 109.218 || 47.232 -> 330.621 MByte/s log_avg of all random - random, method 0 : 0.101 1.470 13.740 43.426 146.812 161.181 || 61.750 -> 432.249 MByte/s - random, method 1 : 0.027 0.429 5.553 35.054 144.442 161.364 || 57.188 -> 400.313 MByte/s - random, method 2 : 0.055 0.832 9.335 39.140 113.341 102.709 || 46.565 -> 325.957 MByte/s log_avg(ring,random) - average, method 0 : 0.102 1.470 13.787 43.307 147.455 161.603 || 62.062 -> 434.431 MByte/s - average, method 1 : 0.026 0.406 5.255 33.057 137.692 161.633 || 55.088 -> 385.617 MByte/s - average, method 2 : 0.055 0.832 9.362 39.320 114.143 105.913 || 46.897 -> 328.281 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 0.715 0.102 0.103 0.101 0.102 0.026 0.055 2 1.425 0.204 0.205 0.203 0.204 0.051 0.110 4 2.696 0.385 0.384 0.387 0.385 0.102 0.212 8 5.352 0.765 0.761 0.768 0.765 0.205 0.423 16 10.293 1.470 1.471 1.470 1.470 0.406 0.832 32 17.567 2.510 2.512 2.507 2.510 0.768 1.522 64 32.507 4.644 4.624 4.663 4.644 1.487 2.826 128 57.240 8.177 8.178 8.177 8.177 2.834 5.265 256 96.507 13.787 13.833 13.740 13.787 5.255 9.362 512 143.616 20.517 20.612 20.421 20.517 8.937 15.098 1024 185.405 26.486 26.507 26.466 26.486 13.960 20.839 2048 186.794 26.685 26.513 26.858 26.685 19.376 24.209 4096 303.149 43.307 43.188 43.426 43.307 33.057 39.320 10624 773.269 110.467 110.653 110.282 110.467 84.576 91.312 27554 917.738 131.105 132.246 129.975 131.105 112.939 106.216 71468 991.149 141.593 142.531 140.661 141.593 129.207 111.883 185364 1032.184 147.455 148.101 146.812 147.455 137.692 114.143 480774 1053.455 150.494 151.717 149.280 150.043 142.586 113.568 1246974 1074.738 153.534 154.319 152.753 153.270 145.728 112.140 3234251 1115.094 159.299 160.061 158.541 158.655 153.609 107.982 8388608 1134.556 162.079 162.244 161.915 161.603 161.633 105.913 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 185364 8388608 -> average -> accumulated p00 ring-3*2&+1 : 0.103 1.405 13.961 43.459 155.197 163.256 -> 64.675 -> 452.724 MByte/s p01 ring-2*4&-1 : 0.103 1.476 13.909 43.135 147.127 162.857 -> 62.096 -> 434.672 MByte/s p02 ring-1*7fix : 0.103 1.486 13.823 43.184 146.056 162.246 -> 61.802 -> 432.613 MByte/s p03 ring-1*7fix : 0.104 1.480 13.793 43.030 146.857 161.245 -> 61.858 -> 433.008 MByte/s p04 ring-1*7fix : 0.103 1.483 13.793 43.449 146.525 162.020 -> 62.096 -> 434.670 MByte/s p05 ring-1*7fix : 0.102 1.500 13.722 42.876 147.042 161.845 -> 62.048 -> 434.336 MByte/s p06 random-cyc-1dim : 0.100 1.507 13.746 43.054 147.287 163.486 -> 62.026 -> 434.183 MByte/s p07 random-cyc-1dim : 0.103 1.491 13.738 43.373 147.859 162.829 -> 62.023 -> 434.159 MByte/s p08 random-cyc-1dim : 0.102 1.499 13.554 43.739 147.318 162.213 -> 62.144 -> 435.008 MByte/s p09 random-cyc-1dim : 0.103 1.482 13.808 42.980 147.825 162.800 -> 62.021 -> 434.149 MByte/s p10 random-cyc-1dim : 0.094 1.442 13.627 43.533 145.446 161.721 -> 61.723 -> 432.059 MByte/s p11 random-cyc-1dim : 0.102 1.482 13.723 43.440 147.585 161.101 -> 61.886 -> 433.202 MByte/s p12 random-cyc-1dim : 0.103 1.496 13.820 43.211 146.281 161.185 -> 61.956 -> 433.689 MByte/s p13 random-cyc-1dim : 0.101 1.476 13.829 43.420 145.804 161.363 -> 61.674 -> 431.717 MByte/s p14 random-cyc-1dim : 0.102 1.479 13.772 43.681 145.846 161.554 -> 61.641 -> 431.485 MByte/s p15 random-cyc-1dim : 0.102 1.347 13.788 43.837 146.891 160.925 -> 61.684 -> 431.789 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.104 1.487 13.758 43.414 146.907 161.767 -> 61.960 -> 433.723 MByte/s p17 best bi-section : 0.121 1.555 9.116 34.484 134.523 149.277 -> 53.886 -> 377.203 MByte/s p18 worst bi-section : 0.121 1.656 9.148 34.541 132.141 147.823 -> 53.340 -> 373.378 MByte/s p19 acyclic-1dim-all : 0.081 1.129 11.489 39.139 131.375 146.491 -> 55.618 -> 389.327 MByte/s p20 acyclic-2dim-all : 0.066 0.963 7.837 30.391 109.171 118.663 -> 44.614 -> 312.299 MByte/s p21 acyclic-3dim-all : 0.066 0.963 7.889 30.320 108.646 120.544 -> 44.640 -> 312.482 MByte/s p22 cyclic-1dim-all : 0.102 1.457 13.604 44.151 144.603 158.165 -> 61.185 -> 428.297 MByte/s p23 cyclic-2dim-all : 0.088 1.262 12.049 38.481 131.205 142.658 -> 54.750 -> 383.253 MByte/s p24 cyclic-3dim-all : 0.088 1.266 12.009 38.846 130.757 141.834 -> 54.811 -> 383.679 MByte/s log_avg of all rings : 0.103 1.471 13.833 43.188 148.101 162.244 || 62.421 -> 436.948 MByte/s log_avg of all random : 0.101 1.470 13.740 43.426 146.812 161.915 || 61.877 -> 433.142 MByte/s log_avg(ring,random) : 0.102 1.470 13.787 43.307 147.455 162.079 || 62.149 -> 435.041 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 435.041 MByte/s on 7 processes ( = 62.149 MByte/s * 7 processes) system parameters : 7 nodes, 1024 MB/node system name: HP-UX hostname : hwwhpv OS release : B.11.00 OS version : A machine : 9000/800 SECTION-BEFF-END b_eff = 435.041 MB/s = 62.149 * 7 PEs with 1024 MB/PE on HP-UX hwwhpv B.11.00 A 9000/800