b_eff = 612.815 MB/s = 76.602 * 8 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E SECTION-HEAD-BEGIN b_eff.c, Revision 3.2 from Nov. 07, 1999 MEMORY_PER_PROCESSOR = 128 MBytes [1M = 1024*1024] 1-dim-paterns: size = 8 1-dim-paterns: size = 8 2-dim-paterns: size = 4 * 2 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 (256), 8192 (128), 16384 (64), 32768 (32), 65536 (16), 131072 (8), 262144 (4), 524288 (2), 1048576 (1), SECTION-HEAD-END SECTION-LOOP-BEGIN measurment loop: elapsed time = 106.739 sec sum of max elapsed time per entries above = 107.844 sec difference = -1.105 sec = 1.0% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-4*2fix => 1 sendrecv_calls with 8 messages, i.e. msgs/used node, all nodes are used p01 ring-2*4fix => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p02 ring-1*8fix => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p03 ring-1*8fix => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p04 ring-1*8fix => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p05 ring-1*8fix => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 8 messages, i.e. msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 8 messages, i.e. msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 14 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 20 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 24 messages, i.e. msgs/used node, all nodes are used p22 cyclic-1dim-all => 2 sendrecv_calls with 16 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 3 sendrecv_calls with 24 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 3 sendrecv_calls with 24 messages, i.e. 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-4*2fix : 72.460 87.968 65.528 -> 87.968 -> 703.744 MByte/s p01 ring-2*4fix : 63.547 87.498 47.829 -> 87.498 -> 699.984 MByte/s p02 ring-1*8fix : 55.617 78.657 47.172 -> 78.657 -> 629.253 MByte/s p03 ring-1*8fix : 54.887 78.961 47.637 -> 78.961 -> 631.684 MByte/s p04 ring-1*8fix : 55.974 78.780 47.059 -> 78.780 -> 630.237 MByte/s p05 ring-1*8fix : 55.752 79.158 47.116 -> 79.158 -> 633.262 MByte/s p06 random-cyc-1dim : 55.827 80.252 47.923 -> 80.252 -> 642.014 MByte/s p07 random-cyc-1dim : 53.745 76.812 45.240 -> 76.812 -> 614.497 MByte/s p08 random-cyc-1dim : 56.690 82.837 47.702 -> 82.837 -> 662.695 MByte/s p09 random-cyc-1dim : 54.033 68.312 46.298 -> 68.312 -> 546.497 MByte/s p10 random-cyc-1dim : 44.120 54.858 40.858 -> 54.858 -> 438.861 MByte/s p11 random-cyc-1dim : 55.646 83.387 47.526 -> 83.387 -> 667.094 MByte/s p12 random-cyc-1dim : 50.347 63.823 41.638 -> 63.823 -> 510.581 MByte/s p13 random-cyc-1dim : 53.018 67.418 45.204 -> 67.418 -> 539.347 MByte/s p14 random-cyc-1dim : 55.187 77.926 48.109 -> 77.926 -> 623.404 MByte/s p15 random-cyc-1dim : 51.895 65.063 44.096 -> 65.063 -> 520.507 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 44.081 52.274 40.458 -> 52.274 -> 418.190 MByte/s p17 best bi-section : 51.790 87.546 50.237 -> 87.546 -> 700.368 MByte/s p18 worst bi-section : 33.495 44.166 33.066 -> 44.166 -> 353.328 MByte/s p19 acyclic-1dim-all : 53.164 67.977 45.626 -> 67.977 -> 543.817 MByte/s p20 acyclic-2dim-all : 49.994 56.127 42.356 -> 56.127 -> 449.019 MByte/s p21 acyclic-3dim-all : 45.114 59.985 44.876 -> 59.985 -> 479.883 MByte/s p22 cyclic-1dim-all : 55.598 78.590 48.367 -> 78.590 -> 628.719 MByte/s p23 cyclic-2dim-all : 57.872 63.547 45.560 -> 63.547 -> 508.379 MByte/s p24 cyclic-3dim-all : 61.516 60.489 47.285 -> 61.516 -> 492.132 MByte/s log_avg of all rings : 59.386 81.733 49.995 || 81.733 -> 653.862 MByte/s log_avg of all random : 52.927 71.478 45.391 || 71.478 -> 571.822 MByte/s log_avg(ring,random) : 56.064 76.433 47.637 ||( 76.433 -> 611.468)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*2fix : 87.640 87.482 87.725 -> 87.725 -> 701.803 MByte/s p01 ring-2*4fix : 87.373 87.071 86.923 -> 87.373 -> 698.987 MByte/s p02 ring-1*8fix : 77.402 77.334 77.711 -> 77.711 -> 621.690 MByte/s p03 ring-1*8fix : 78.197 77.971 78.381 -> 78.381 -> 627.049 MByte/s p04 ring-1*8fix : 75.345 78.073 77.559 -> 78.073 -> 624.588 MByte/s p05 ring-1*8fix : 78.165 77.726 78.376 -> 78.376 -> 627.005 MByte/s p06 random-cyc-1dim : 78.751 77.965 78.918 -> 78.918 -> 631.343 MByte/s p07 random-cyc-1dim : 75.646 76.261 75.656 -> 76.261 -> 610.087 MByte/s p08 random-cyc-1dim : 81.593 81.424 81.166 -> 81.593 -> 652.742 MByte/s p09 random-cyc-1dim : 66.730 67.872 67.837 -> 67.872 -> 542.975 MByte/s p10 random-cyc-1dim : 54.152 54.474 54.914 -> 54.914 -> 439.316 MByte/s p11 random-cyc-1dim : 82.796 82.582 82.154 -> 82.796 -> 662.372 MByte/s p12 random-cyc-1dim : 63.304 63.220 63.476 -> 63.476 -> 507.809 MByte/s p13 random-cyc-1dim : 66.580 67.322 66.350 -> 67.322 -> 538.578 MByte/s p14 random-cyc-1dim : 76.572 77.840 77.007 -> 77.840 -> 622.717 MByte/s p15 random-cyc-1dim : 64.565 64.809 64.774 -> 64.809 -> 518.470 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 52.812 52.750 52.382 -> 52.812 -> 422.494 MByte/s p17 best bi-section : 87.086 87.467 87.297 -> 87.467 -> 699.734 MByte/s p18 worst bi-section : 44.134 43.969 44.044 -> 44.134 -> 353.072 MByte/s p19 acyclic-1dim-all : 68.124 68.508 68.098 -> 68.508 -> 548.061 MByte/s p20 acyclic-2dim-all : 60.887 61.684 59.963 -> 61.684 -> 493.472 MByte/s p21 acyclic-3dim-all : 60.584 60.521 60.604 -> 60.604 -> 484.832 MByte/s p22 cyclic-1dim-all : 77.338 78.161 77.213 -> 78.161 -> 625.288 MByte/s p23 cyclic-2dim-all : 65.679 66.097 66.331 -> 66.331 -> 530.647 MByte/s p24 cyclic-3dim-all : 66.417 66.813 63.194 -> 66.813 -> 534.503 MByte/s log_avg of all rings : 80.541 80.822 80.996 || 81.155 -> 649.238 MByte/s log_avg of all random : 70.493 70.820 70.690 || 71.025 -> 568.201 MByte/s log_avg(ring,random) : 75.350 75.656 75.667 ||( 75.921 -> 607.370)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 65536 1048576 -> average -> accumulated p00 ring-4*2fix p00 method 0 : 0.056 0.885 10.423 52.298 179.428 208.453 -> 72.460 -> 579.680 MByte/s p00 method 1 : 0.054 1.046 14.710 114.961 200.020 209.840 -> 87.968 -> 703.744 MByte/s p00 method 2 : 0.025 0.403 5.362 27.342 168.870 207.001 -> 65.528 -> 524.223 MByte/s p01 ring-2*4fix p01 method 0 : 0.055 0.869 10.230 50.318 172.701 154.366 -> 63.547 -> 508.376 MByte/s p01 method 1 : 0.070 1.311 17.965 120.928 194.473 199.938 -> 87.498 -> 699.984 MByte/s p01 method 2 : 0.024 0.364 4.949 29.885 119.942 111.232 -> 47.829 -> 382.628 MByte/s p02 ring-1*8fix p02 method 0 : 0.053 0.839 9.900 49.059 139.785 145.286 -> 55.617 -> 444.936 MByte/s p02 method 1 : 0.067 1.290 17.511 119.442 161.491 191.759 -> 78.657 -> 629.253 MByte/s p02 method 2 : 0.022 0.350 4.551 30.088 121.435 131.579 -> 47.172 -> 377.376 MByte/s p03 ring-1*8fix p03 method 0 : 0.053 0.857 9.761 49.642 138.241 141.717 -> 54.887 -> 439.097 MByte/s p03 method 1 : 0.068 1.287 17.492 119.373 159.894 186.966 -> 78.961 -> 631.684 MByte/s p03 method 2 : 0.022 0.347 4.561 30.380 119.649 142.080 -> 47.637 -> 381.098 MByte/s p04 ring-1*8fix p04 method 0 : 0.054 0.838 10.034 50.075 137.298 142.774 -> 55.974 -> 447.791 MByte/s p04 method 1 : 0.067 1.277 17.519 119.628 159.477 192.306 -> 78.780 -> 630.237 MByte/s p04 method 2 : 0.022 0.352 4.509 30.529 122.505 130.224 -> 47.059 -> 376.473 MByte/s p05 ring-1*8fix p05 method 0 : 0.053 0.855 9.990 50.140 135.611 143.054 -> 55.752 -> 446.020 MByte/s p05 method 1 : 0.068 1.282 17.463 118.889 158.989 195.204 -> 79.158 -> 633.262 MByte/s p05 method 2 : 0.022 0.348 4.612 30.429 119.571 128.588 -> 47.116 -> 376.929 MByte/s p06 random-cyc-1dim p06 method 0 : 0.055 0.856 10.253 50.366 138.462 138.821 -> 55.827 -> 446.620 MByte/s p06 method 1 : 0.068 1.296 17.539 119.806 162.581 193.879 -> 80.252 -> 642.014 MByte/s p06 method 2 : 0.022 0.349 4.578 30.429 126.866 131.789 -> 47.923 -> 383.386 MByte/s p07 random-cyc-1dim p07 method 0 : 0.053 0.847 9.843 49.485 135.274 127.283 -> 53.745 -> 429.962 MByte/s p07 method 1 : 0.069 1.296 17.637 115.069 171.022 167.792 -> 76.812 -> 614.497 MByte/s p07 method 2 : 0.022 0.348 4.538 30.012 118.179 111.185 -> 45.240 -> 361.918 MByte/s p08 random-cyc-1dim p08 method 0 : 0.052 0.842 10.109 50.113 138.825 159.126 -> 56.690 -> 453.519 MByte/s p08 method 1 : 0.068 1.297 17.663 121.147 169.666 192.479 -> 82.837 -> 662.695 MByte/s p08 method 2 : 0.022 0.352 4.574 30.315 125.731 126.077 -> 47.702 -> 381.612 MByte/s p09 random-cyc-1dim p09 method 0 : 0.054 0.841 9.906 50.278 133.456 139.298 -> 54.033 -> 432.262 MByte/s p09 method 1 : 0.071 1.295 17.582 108.854 139.311 151.127 -> 68.312 -> 546.497 MByte/s p09 method 2 : 0.022 0.347 4.538 30.271 113.319 155.193 -> 46.298 -> 370.387 MByte/s p10 random-cyc-1dim p10 method 0 : 0.053 0.851 9.933 46.895 107.155 100.444 -> 44.120 -> 352.961 MByte/s p10 method 1 : 0.070 1.264 17.433 98.652 112.219 104.995 -> 54.858 -> 438.861 MByte/s p10 method 2 : 0.022 0.354 4.523 29.879 100.734 114.071 -> 40.858 -> 326.866 MByte/s p11 random-cyc-1dim p11 method 0 : 0.053 0.852 9.802 50.049 136.590 146.620 -> 55.646 -> 445.166 MByte/s p11 method 1 : 0.069 1.276 17.658 120.856 183.393 185.681 -> 83.387 -> 667.094 MByte/s p11 method 2 : 0.022 0.348 4.546 30.252 121.835 144.999 -> 47.526 -> 380.209 MByte/s p12 random-cyc-1dim p12 method 0 : 0.052 0.854 10.064 49.024 120.077 126.861 -> 50.347 -> 402.773 MByte/s p12 method 1 : 0.071 1.291 17.496 99.246 133.104 134.989 -> 63.823 -> 510.581 MByte/s p12 method 2 : 0.022 0.341 4.446 30.148 108.556 106.815 -> 41.638 -> 333.101 MByte/s p13 random-cyc-1dim p13 method 0 : 0.054 0.835 10.150 49.489 131.955 137.803 -> 53.018 -> 424.146 MByte/s p13 method 1 : 0.070 1.282 17.471 101.745 141.778 145.844 -> 67.418 -> 539.347 MByte/s p13 method 2 : 0.022 0.380 4.524 30.179 114.988 113.237 -> 45.204 -> 361.629 MByte/s p14 random-cyc-1dim p14 method 0 : 0.053 0.844 10.175 50.497 135.566 140.424 -> 55.187 -> 441.495 MByte/s p14 method 1 : 0.071 1.286 17.523 117.774 157.604 190.063 -> 77.926 -> 623.404 MByte/s p14 method 2 : 0.022 0.358 4.576 30.163 122.354 140.290 -> 48.109 -> 384.872 MByte/s p15 random-cyc-1dim p15 method 0 : 0.054 0.848 9.918 49.353 125.321 125.923 -> 51.895 -> 415.159 MByte/s p15 method 1 : 0.070 1.295 17.525 100.348 137.781 137.495 -> 65.063 -> 520.507 MByte/s p15 method 2 : 0.022 0.346 4.529 30.044 110.065 119.502 -> 44.096 -> 352.767 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.053 0.840 9.416 47.828 103.597 109.176 -> 44.081 -> 352.648 MByte/s p16 method 1 : 0.070 1.275 17.244 88.041 102.530 103.441 -> 52.274 -> 418.190 MByte/s p16 method 2 : 0.022 0.348 4.451 30.108 101.326 115.673 -> 40.458 -> 323.661 MByte/s p17 best bi-section p17 method 0 : 0.034 0.518 5.734 28.506 131.771 161.877 -> 51.790 -> 414.319 MByte/s p17 method 1 : 0.057 1.045 14.261 113.715 199.503 210.166 -> 87.546 -> 700.368 MByte/s p17 method 2 : 0.015 0.241 3.534 28.095 128.472 160.992 -> 50.237 -> 401.896 MByte/s p18 worst bi-section p18 method 0 : 0.027 0.402 5.068 27.466 81.011 94.806 -> 33.495 -> 267.959 MByte/s p18 method 1 : 0.052 0.988 13.819 70.318 80.168 115.791 -> 44.166 -> 353.328 MByte/s p18 method 2 : 0.015 0.238 3.584 27.230 83.053 94.821 -> 33.066 -> 264.530 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.045 0.718 8.593 44.147 127.726 141.493 -> 53.164 -> 425.309 MByte/s p19 method 1 : 0.064 1.157 15.833 102.795 150.900 129.690 -> 67.977 -> 543.817 MByte/s p19 method 2 : 0.019 0.299 3.945 27.981 113.782 122.156 -> 45.626 -> 365.005 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.032 0.514 5.862 25.102 126.950 155.583 -> 49.994 -> 399.950 MByte/s p20 method 1 : 0.066 1.238 16.224 83.509 118.371 123.666 -> 56.127 -> 449.019 MByte/s p20 method 2 : 0.016 0.255 3.369 24.578 109.690 115.158 -> 42.356 -> 338.851 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.029 0.456 5.500 28.164 113.427 135.477 -> 45.114 -> 360.912 MByte/s p21 method 1 : 0.078 1.444 18.528 94.278 122.063 128.616 -> 59.985 -> 479.883 MByte/s p21 method 2 : 0.015 0.237 3.218 25.933 115.712 130.496 -> 44.876 -> 359.011 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.056 0.858 9.918 50.729 134.738 140.281 -> 55.598 -> 444.788 MByte/s p22 method 1 : 0.066 1.202 16.519 116.716 161.794 192.816 -> 78.590 -> 628.719 MByte/s p22 method 2 : 0.022 0.350 4.580 30.306 121.810 138.827 -> 48.367 -> 386.939 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.056 0.858 10.269 51.341 146.550 139.825 -> 57.872 -> 462.978 MByte/s p23 method 1 : 0.074 1.356 18.348 99.449 131.592 137.875 -> 63.547 -> 508.379 MByte/s p23 method 2 : 0.022 0.345 4.487 30.370 116.728 107.228 -> 45.560 -> 364.480 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.056 0.871 10.293 51.290 153.781 146.580 -> 61.516 -> 492.132 MByte/s p24 method 1 : 0.074 1.345 18.058 94.386 121.989 127.944 -> 60.489 -> 483.911 MByte/s p24 method 2 : 0.022 0.348 4.498 30.673 118.379 131.620 -> 47.285 -> 378.281 MByte/s log_avg of all rings - ring, method 0 : 0.054 0.857 10.054 50.245 149.465 154.369 || 59.386 -> 475.091 MByte/s - ring, method 1 : 0.065 1.245 17.073 118.855 171.526 195.868 || 81.733 -> 653.862 MByte/s - ring, method 2 : 0.023 0.360 4.748 29.754 127.573 138.985 || 49.995 -> 399.960 MByte/s log_avg of all random - random, method 0 : 0.053 0.847 10.014 49.545 129.893 133.388 || 52.927 -> 423.416 MByte/s - random, method 1 : 0.070 1.288 17.553 109.970 149.407 157.664 || 71.478 -> 571.822 MByte/s - random, method 2 : 0.022 0.352 4.537 30.169 115.989 125.401 || 45.391 -> 363.125 MByte/s log_avg(ring,random) - average, method 0 : 0.054 0.852 10.034 49.894 139.336 143.495 || 56.064 -> 448.510 MByte/s - average, method 1 : 0.068 1.266 17.311 114.327 160.085 175.731 || 76.433 -> 611.468 MByte/s - average, method 2 : 0.022 0.356 4.641 29.961 121.643 132.018 || 47.637 -> 381.097 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 0.542 0.068 0.066 0.070 0.054 0.068 0.022 2 1.090 0.136 0.132 0.140 0.108 0.136 0.045 4 2.680 0.335 0.329 0.341 0.216 0.335 0.090 8 5.335 0.667 0.653 0.681 0.430 0.667 0.179 16 10.130 1.266 1.245 1.288 0.852 1.266 0.356 32 19.451 2.431 2.393 2.471 1.394 2.431 0.625 64 35.236 4.404 4.337 4.472 2.551 4.404 1.181 128 77.396 9.674 9.525 9.827 5.528 9.674 2.451 256 138.488 17.311 17.073 17.553 10.034 17.311 4.641 512 248.418 31.052 30.773 31.334 17.563 31.052 8.470 1024 424.671 53.084 52.928 53.240 28.066 53.084 14.613 2048 642.915 80.364 81.684 79.066 37.845 80.364 22.424 4096 914.613 114.327 118.855 109.970 49.894 114.327 29.961 8192 1114.535 139.317 150.182 129.237 72.000 139.317 66.912 16384 1244.434 155.554 170.800 141.669 98.578 155.554 92.235 32768 1249.245 156.156 165.675 147.183 123.236 156.156 110.749 65536 1280.679 160.085 171.526 149.407 139.336 160.085 121.643 131072 1300.627 162.578 171.759 153.888 147.344 162.048 128.126 262144 1331.413 166.427 179.528 154.281 148.304 165.109 131.414 524288 1390.897 173.862 189.028 159.913 149.085 172.800 130.421 1048576 1413.565 176.696 195.868 159.400 143.495 175.731 132.018 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 65536 1048576 -> average -> accumulated p00 ring-4*2fix : 0.056 1.046 14.710 114.961 200.020 209.840 -> 87.968 -> 703.746 MByte/s p01 ring-2*4fix : 0.070 1.311 17.965 120.928 194.473 199.938 -> 87.498 -> 699.984 MByte/s p02 ring-1*8fix : 0.067 1.290 17.511 119.442 161.491 191.759 -> 78.657 -> 629.253 MByte/s p03 ring-1*8fix : 0.068 1.287 17.492 119.373 159.894 186.966 -> 78.961 -> 631.684 MByte/s p04 ring-1*8fix : 0.067 1.277 17.519 119.628 159.477 192.306 -> 78.780 -> 630.237 MByte/s p05 ring-1*8fix : 0.068 1.282 17.463 118.889 158.989 195.204 -> 79.158 -> 633.262 MByte/s p06 random-cyc-1dim : 0.068 1.296 17.539 119.806 162.581 193.879 -> 80.252 -> 642.014 MByte/s p07 random-cyc-1dim : 0.069 1.296 17.637 115.069 171.022 167.792 -> 76.812 -> 614.497 MByte/s p08 random-cyc-1dim : 0.068 1.297 17.663 121.147 169.666 192.479 -> 82.837 -> 662.695 MByte/s p09 random-cyc-1dim : 0.071 1.295 17.582 108.854 139.311 155.193 -> 69.879 -> 559.028 MByte/s p10 random-cyc-1dim : 0.070 1.264 17.433 98.652 112.219 114.071 -> 55.723 -> 445.783 MByte/s p11 random-cyc-1dim : 0.069 1.276 17.658 120.856 183.393 185.681 -> 83.387 -> 667.094 MByte/s p12 random-cyc-1dim : 0.071 1.291 17.496 99.246 133.104 134.989 -> 63.823 -> 510.581 MByte/s p13 random-cyc-1dim : 0.070 1.282 17.471 101.745 141.778 145.844 -> 67.418 -> 539.347 MByte/s p14 random-cyc-1dim : 0.071 1.286 17.523 117.774 157.604 190.063 -> 77.926 -> 623.404 MByte/s p15 random-cyc-1dim : 0.070 1.295 17.525 100.348 137.781 137.495 -> 65.436 -> 523.485 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.070 1.275 17.244 88.041 103.597 115.673 -> 53.402 -> 427.218 MByte/s p17 best bi-section : 0.057 1.045 14.261 113.715 199.503 210.166 -> 87.546 -> 700.368 MByte/s p18 worst bi-section : 0.052 0.988 13.819 70.318 83.053 115.791 -> 44.442 -> 355.536 MByte/s p19 acyclic-1dim-all : 0.064 1.157 15.833 102.795 150.900 141.493 -> 69.119 -> 552.952 MByte/s p20 acyclic-2dim-all : 0.066 1.238 16.224 83.509 126.950 155.583 -> 61.960 -> 495.678 MByte/s p21 acyclic-3dim-all : 0.078 1.444 18.528 94.278 122.063 135.477 -> 60.761 -> 486.090 MByte/s p22 cyclic-1dim-all : 0.066 1.202 16.519 116.716 161.794 192.816 -> 78.590 -> 628.719 MByte/s p23 cyclic-2dim-all : 0.074 1.356 18.348 99.449 146.550 139.825 -> 67.318 -> 538.545 MByte/s p24 cyclic-3dim-all : 0.074 1.345 18.058 94.386 153.781 146.580 -> 69.307 -> 554.458 MByte/s log_avg of all rings : 0.066 1.245 17.073 118.855 171.526 195.868 || 81.733 -> 653.862 MByte/s log_avg of all random : 0.070 1.288 17.553 109.970 149.407 159.400 || 71.793 -> 574.345 MByte/s log_avg(ring,random) : 0.068 1.266 17.311 114.327 160.085 176.696 || 76.602 -> 612.815 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 612.815 MByte/s on 8 processes ( = 76.602 MByte/s * 8 processes) system parameters : 8 nodes, 128 MB/node system name: sn6715 hostname : hwwt3e OS release : 2.0.4.71 OS version : unicosmk machine : CRAY T3E SECTION-BEFF-END b_eff = 612.815 MB/s = 76.602 * 8 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E