b_eff = 2725.891 MB/s = 56.789 * 48 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 = 48 1-dim-paterns: size = 48 2-dim-paterns: size = 8 * 6 3-dim-paterns: size = 4 * 4 * 3 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 = 152.788 sec sum of max elapsed time per entries above = 155.701 sec difference = -2.913 sec = 1.9% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-24*2fix => 1 sendrecv_calls with 48 messages, i.e. msgs/used node, all nodes are used p01 ring-12*4fix => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p02 ring-6*8fix => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p03 ring-3*16fix => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p04 ring-1*48fix => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p05 ring-1*48fix => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 48 messages, i.e. msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 48 messages, i.e. msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 94 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 164 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 208 messages, i.e. msgs/used node, all nodes are used p22 cyclic-1dim-all => 2 sendrecv_calls with 96 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 192 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 6 sendrecv_calls with 288 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-24*2fix : 71.931 73.300 59.285 -> 73.300 -> 3518.409 MByte/s p01 ring-12*4fix : 58.712 77.509 45.356 -> 77.509 -> 3720.426 MByte/s p02 ring-6*8fix : 53.172 69.215 45.017 -> 69.215 -> 3322.342 MByte/s p03 ring-3*16fix : 51.990 66.841 43.657 -> 66.841 -> 3208.370 MByte/s p04 ring-1*48fix : 52.219 63.008 44.470 -> 63.008 -> 3024.378 MByte/s p05 ring-1*48fix : 52.021 63.439 44.815 -> 63.439 -> 3045.055 MByte/s p06 random-cyc-1dim : 42.322 46.418 36.482 -> 46.418 -> 2228.069 MByte/s p07 random-cyc-1dim : 39.470 43.032 34.107 -> 43.032 -> 2065.548 MByte/s p08 random-cyc-1dim : 42.729 48.420 37.383 -> 48.420 -> 2324.161 MByte/s p09 random-cyc-1dim : 44.486 49.898 37.965 -> 49.898 -> 2395.112 MByte/s p10 random-cyc-1dim : 34.337 36.538 30.441 -> 36.538 -> 1753.831 MByte/s p11 random-cyc-1dim : 41.329 44.888 36.668 -> 44.888 -> 2154.617 MByte/s p12 random-cyc-1dim : 40.124 43.643 35.384 -> 43.643 -> 2094.854 MByte/s p13 random-cyc-1dim : 42.420 46.733 37.597 -> 46.733 -> 2243.171 MByte/s p14 random-cyc-1dim : 40.730 43.517 35.137 -> 43.517 -> 2088.823 MByte/s p15 random-cyc-1dim : 36.752 39.820 32.041 -> 39.820 -> 1911.384 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 19.324 20.398 17.778 -> 20.398 -> 979.106 MByte/s p17 best bi-section : 51.545 73.309 49.074 -> 73.309 -> 3518.827 MByte/s p18 worst bi-section : 18.172 21.117 17.031 -> 21.117 -> 1013.640 MByte/s p19 acyclic-1dim-all : 54.634 67.999 45.640 -> 67.999 -> 3263.967 MByte/s p20 acyclic-2dim-all : 43.607 50.117 36.107 -> 50.117 -> 2405.620 MByte/s p21 acyclic-3dim-all : 31.201 35.520 28.046 -> 35.520 -> 1704.937 MByte/s p22 cyclic-1dim-all : 51.691 62.831 44.245 -> 62.831 -> 3015.882 MByte/s p23 cyclic-2dim-all : 44.580 49.595 37.739 -> 49.595 -> 2380.549 MByte/s p24 cyclic-3dim-all : 38.178 40.690 32.374 -> 40.690 -> 1953.124 MByte/s log_avg of all rings : 56.261 68.692 46.819 || 68.692 -> 3297.235 MByte/s log_avg of all random : 40.364 44.124 35.238 || 44.124 -> 2117.961 MByte/s log_avg(ring,random) : 47.654 55.054 40.618 ||( 55.054 -> 2642.615)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-24*2fix : 74.092 74.303 73.584 -> 74.303 -> 3566.531 MByte/s p01 ring-12*4fix : 76.935 77.158 75.790 -> 77.158 -> 3703.596 MByte/s p02 ring-6*8fix : 68.451 68.711 68.252 -> 68.711 -> 3298.145 MByte/s p03 ring-3*16fix : 66.631 63.746 65.907 -> 66.631 -> 3198.282 MByte/s p04 ring-1*48fix : 62.369 63.022 63.016 -> 63.022 -> 3025.074 MByte/s p05 ring-1*48fix : 62.803 62.051 62.433 -> 62.803 -> 3014.553 MByte/s p06 random-cyc-1dim : 48.074 47.501 48.079 -> 48.079 -> 2307.784 MByte/s p07 random-cyc-1dim : 44.735 45.352 44.990 -> 45.352 -> 2176.914 MByte/s p08 random-cyc-1dim : 48.953 48.730 49.790 -> 49.790 -> 2389.920 MByte/s p09 random-cyc-1dim : 50.555 50.670 50.338 -> 50.670 -> 2432.167 MByte/s p10 random-cyc-1dim : 38.151 38.396 38.703 -> 38.703 -> 1857.745 MByte/s p11 random-cyc-1dim : 46.473 46.576 46.271 -> 46.576 -> 2235.663 MByte/s p12 random-cyc-1dim : 45.726 45.799 45.568 -> 45.799 -> 2198.343 MByte/s p13 random-cyc-1dim : 48.095 48.916 48.802 -> 48.916 -> 2347.951 MByte/s p14 random-cyc-1dim : 44.858 44.893 46.120 -> 46.120 -> 2213.779 MByte/s p15 random-cyc-1dim : 41.763 41.751 41.635 -> 41.763 -> 2004.648 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 21.377 20.941 20.934 -> 21.377 -> 1026.104 MByte/s p17 best bi-section : 72.931 72.463 72.971 -> 72.971 -> 3502.631 MByte/s p18 worst bi-section : 21.068 21.004 21.024 -> 21.068 -> 1011.262 MByte/s p19 acyclic-1dim-all : 67.416 66.416 66.641 -> 67.416 -> 3235.992 MByte/s p20 acyclic-2dim-all : 52.678 52.332 52.522 -> 52.678 -> 2528.522 MByte/s p21 acyclic-3dim-all : 37.632 37.043 37.240 -> 37.632 -> 1806.327 MByte/s p22 cyclic-1dim-all : 62.103 62.837 61.452 -> 62.837 -> 3016.180 MByte/s p23 cyclic-2dim-all : 52.387 52.146 51.866 -> 52.387 -> 2514.559 MByte/s p24 cyclic-3dim-all : 44.870 44.996 44.937 -> 44.996 -> 2159.794 MByte/s log_avg of all rings : 68.336 67.923 67.981 || 68.564 -> 3291.060 MByte/s log_avg of all random : 45.601 45.724 45.895 || 46.040 -> 2209.918 MByte/s log_avg(ring,random) : 55.823 55.729 55.856 ||( 56.184 -> 2696.845)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-24*2fix p00 method 0 : 0.055 0.871 9.934 51.249 179.751 206.870 -> 71.931 -> 3452.711 MByte/s p00 method 1 : 0.021 0.364 5.557 63.733 183.911 207.573 -> 73.300 -> 3518.409 MByte/s p00 method 2 : 0.025 0.395 5.332 26.125 168.491 163.275 -> 59.285 -> 2845.672 MByte/s p01 ring-12*4fix p01 method 0 : 0.054 0.848 9.986 50.267 139.540 153.699 -> 58.712 -> 2818.158 MByte/s p01 method 1 : 0.035 0.601 8.904 86.198 187.129 196.232 -> 77.509 -> 3720.426 MByte/s p01 method 2 : 0.022 0.342 4.664 29.522 119.009 113.345 -> 45.356 -> 2177.081 MByte/s p02 ring-6*8fix p02 method 0 : 0.052 0.825 9.475 48.156 134.239 132.077 -> 53.172 -> 2552.276 MByte/s p02 method 1 : 0.034 0.597 8.896 80.990 154.503 187.284 -> 69.215 -> 3322.342 MByte/s p02 method 2 : 0.022 0.341 4.479 29.599 116.131 117.973 -> 45.017 -> 2160.823 MByte/s p03 ring-3*16fix p03 method 0 : 0.049 0.779 8.567 41.729 129.543 138.667 -> 51.990 -> 2495.501 MByte/s p03 method 1 : 0.034 0.593 8.817 82.250 151.119 173.853 -> 66.841 -> 3208.370 MByte/s p03 method 2 : 0.022 0.342 4.447 29.534 115.211 112.022 -> 43.657 -> 2095.556 MByte/s p04 ring-1*48fix p04 method 0 : 0.047 0.752 8.401 38.061 130.640 131.170 -> 52.219 -> 2506.495 MByte/s p04 method 1 : 0.034 0.588 8.709 83.663 138.655 161.325 -> 63.008 -> 3024.378 MByte/s p04 method 2 : 0.022 0.343 4.457 29.894 108.477 124.994 -> 44.470 -> 2134.567 MByte/s p05 ring-1*48fix p05 method 0 : 0.047 0.751 8.348 38.113 131.637 130.964 -> 52.021 -> 2496.997 MByte/s p05 method 1 : 0.034 0.589 8.769 83.245 139.158 160.268 -> 63.439 -> 3045.055 MByte/s p05 method 2 : 0.022 0.343 4.448 29.797 117.995 118.046 -> 44.815 -> 2151.127 MByte/s p06 random-cyc-1dim p06 method 0 : 0.044 0.701 7.910 36.832 105.392 103.575 -> 42.322 -> 2031.443 MByte/s p06 method 1 : 0.034 0.595 8.713 76.987 104.830 89.049 -> 46.418 -> 2228.069 MByte/s p06 method 2 : 0.021 0.333 4.311 29.292 90.467 88.815 -> 36.482 -> 1751.142 MByte/s p07 random-cyc-1dim p07 method 0 : 0.045 0.705 8.156 37.515 92.043 93.111 -> 39.470 -> 1894.551 MByte/s p07 method 1 : 0.033 0.570 8.481 76.094 97.888 71.330 -> 43.032 -> 2065.548 MByte/s p07 method 2 : 0.021 0.336 4.350 29.295 86.820 76.500 -> 34.107 -> 1637.128 MByte/s p08 random-cyc-1dim p08 method 0 : 0.046 0.734 7.902 36.908 109.001 87.632 -> 42.729 -> 2051.004 MByte/s p08 method 1 : 0.034 0.590 8.778 77.846 109.711 77.559 -> 48.420 -> 2324.161 MByte/s p08 method 2 : 0.021 0.336 4.354 29.386 92.882 96.468 -> 37.383 -> 1794.367 MByte/s p09 random-cyc-1dim p09 method 0 : 0.045 0.726 8.014 37.856 112.509 109.367 -> 44.486 -> 2135.304 MByte/s p09 method 1 : 0.034 0.589 8.776 78.384 114.250 98.252 -> 49.898 -> 2395.112 MByte/s p09 method 2 : 0.021 0.336 4.400 29.432 96.970 92.842 -> 37.965 -> 1822.325 MByte/s p10 random-cyc-1dim p10 method 0 : 0.044 0.714 7.887 37.201 83.517 71.474 -> 34.337 -> 1648.172 MByte/s p10 method 1 : 0.035 0.598 8.848 70.982 76.095 59.414 -> 36.538 -> 1753.831 MByte/s p10 method 2 : 0.021 0.334 4.308 29.164 76.407 61.754 -> 30.441 -> 1461.152 MByte/s p11 random-cyc-1dim p11 method 0 : 0.046 0.719 8.086 36.836 103.073 92.701 -> 41.329 -> 1983.804 MByte/s p11 method 1 : 0.034 0.581 8.625 75.184 97.477 89.095 -> 44.888 -> 2154.617 MByte/s p11 method 2 : 0.021 0.337 4.396 29.288 93.376 89.747 -> 36.668 -> 1760.068 MByte/s p12 random-cyc-1dim p12 method 0 : 0.045 0.705 7.768 36.790 101.690 89.659 -> 40.124 -> 1925.958 MByte/s p12 method 1 : 0.034 0.593 8.769 75.504 95.966 77.204 -> 43.643 -> 2094.854 MByte/s p12 method 2 : 0.021 0.334 4.322 29.398 88.151 87.328 -> 35.384 -> 1698.435 MByte/s p13 random-cyc-1dim p13 method 0 : 0.044 0.713 7.840 37.108 107.587 96.201 -> 42.420 -> 2036.164 MByte/s p13 method 1 : 0.034 0.588 8.764 77.036 102.771 85.066 -> 46.733 -> 2243.171 MByte/s p13 method 2 : 0.021 0.333 4.300 29.311 97.902 93.632 -> 37.597 -> 1804.680 MByte/s p14 random-cyc-1dim p14 method 0 : 0.045 0.735 7.851 38.368 98.935 93.622 -> 40.730 -> 1955.053 MByte/s p14 method 1 : 0.034 0.584 8.680 78.137 94.196 85.596 -> 43.517 -> 2088.823 MByte/s p14 method 2 : 0.021 0.337 4.377 29.155 86.371 84.392 -> 35.137 -> 1686.565 MByte/s p15 random-cyc-1dim p15 method 0 : 0.046 0.738 7.946 36.846 93.238 68.185 -> 36.752 -> 1764.103 MByte/s p15 method 1 : 0.035 0.594 8.802 72.854 86.342 60.096 -> 39.820 -> 1911.384 MByte/s p15 method 2 : 0.021 0.337 4.395 29.436 83.681 69.588 -> 32.041 -> 1537.985 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.042 0.691 7.566 30.768 40.644 28.194 -> 19.324 -> 927.546 MByte/s p16 method 1 : 0.034 0.591 8.790 39.178 39.520 27.675 -> 20.398 -> 979.106 MByte/s p16 method 2 : 0.021 0.333 4.237 27.092 37.697 38.559 -> 17.778 -> 853.368 MByte/s p17 best bi-section p17 method 0 : 0.034 0.497 5.467 27.978 130.949 162.695 -> 51.545 -> 2474.145 MByte/s p17 method 1 : 0.022 0.361 5.510 62.938 183.294 209.053 -> 73.309 -> 3518.827 MByte/s p17 method 2 : 0.015 0.238 3.226 24.337 126.377 160.992 -> 49.074 -> 2355.541 MByte/s p18 worst bi-section p18 method 0 : 0.024 0.369 4.475 24.931 36.438 41.931 -> 18.172 -> 872.249 MByte/s p18 method 1 : 0.021 0.350 5.325 33.940 38.525 56.004 -> 21.117 -> 1013.640 MByte/s p18 method 2 : 0.015 0.233 3.325 22.855 36.528 36.906 -> 17.031 -> 817.485 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.047 0.750 8.142 38.298 142.286 139.616 -> 54.634 -> 2622.439 MByte/s p19 method 1 : 0.035 0.590 8.746 84.786 166.504 137.504 -> 67.999 -> 3263.967 MByte/s p19 method 2 : 0.021 0.333 4.371 29.540 122.358 110.835 -> 45.640 -> 2190.734 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.035 0.547 5.999 28.526 112.081 118.502 -> 43.607 -> 2093.118 MByte/s p20 method 1 : 0.044 0.772 10.938 80.087 110.397 97.194 -> 50.117 -> 2405.620 MByte/s p20 method 2 : 0.018 0.292 3.782 25.791 95.821 88.089 -> 36.107 -> 1733.152 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.030 0.470 5.285 25.219 77.929 76.579 -> 31.201 -> 1497.641 MByte/s p21 method 1 : 0.045 0.802 11.003 61.917 71.923 63.190 -> 35.520 -> 1704.937 MByte/s p21 method 2 : 0.015 0.246 3.187 21.581 72.342 60.991 -> 28.046 -> 1346.208 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.047 0.757 8.359 38.381 129.604 133.947 -> 51.691 -> 2481.165 MByte/s p22 method 1 : 0.033 0.571 8.483 82.202 140.508 161.712 -> 62.831 -> 3015.882 MByte/s p22 method 2 : 0.022 0.345 4.469 29.848 113.814 112.292 -> 44.245 -> 2123.780 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.040 0.623 6.815 32.905 113.551 115.561 -> 44.580 -> 2139.854 MByte/s p23 method 1 : 0.049 0.847 12.160 80.294 107.954 95.730 -> 49.595 -> 2380.549 MByte/s p23 method 2 : 0.022 0.340 4.392 29.628 97.380 92.721 -> 37.739 -> 1811.482 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.038 0.589 6.501 31.140 95.057 95.803 -> 38.178 -> 1832.533 MByte/s p24 method 1 : 0.058 1.012 14.202 70.923 80.931 76.300 -> 40.690 -> 1953.124 MByte/s p24 method 2 : 0.021 0.333 4.251 28.831 80.295 72.160 -> 32.374 -> 1553.966 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.803 9.092 44.252 139.911 146.789 || 56.261 -> 2700.532 MByte/s - ring, method 1 : 0.031 0.547 8.165 79.626 157.909 180.245 || 68.692 -> 3297.235 MByte/s - ring, method 2 : 0.022 0.351 4.627 29.047 122.829 123.849 || 46.819 -> 2247.289 MByte/s log_avg of all random - random, method 0 : 0.045 0.719 7.935 37.223 100.330 89.699 || 40.364 -> 1937.484 MByte/s - random, method 1 : 0.034 0.588 8.723 75.866 97.360 78.295 || 44.124 -> 2117.961 MByte/s - random, method 2 : 0.021 0.335 4.351 29.315 89.085 83.356 || 35.238 -> 1691.442 MByte/s log_avg(ring,random) - average, method 0 : 0.048 0.760 8.494 40.586 118.479 114.747 || 47.654 -> 2287.408 MByte/s - average, method 1 : 0.033 0.567 8.440 77.723 123.992 118.795 || 55.054 -> 2642.615 MByte/s - average, method 2 : 0.022 0.343 4.487 29.181 104.605 101.605 || 40.618 -> 1949.656 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 2.293 0.048 0.051 0.045 0.048 0.033 0.022 2 4.583 0.095 0.102 0.090 0.095 0.066 0.043 4 9.161 0.191 0.203 0.180 0.191 0.145 0.086 8 18.355 0.382 0.406 0.361 0.382 0.290 0.173 16 36.475 0.760 0.803 0.719 0.760 0.567 0.343 32 58.146 1.211 1.272 1.153 1.161 1.115 0.601 64 109.676 2.285 2.371 2.202 2.109 2.131 1.144 128 230.920 4.811 5.038 4.594 4.612 4.440 2.369 256 431.538 8.990 9.266 8.723 8.494 8.440 4.487 512 801.890 16.706 16.889 16.525 14.946 16.006 8.174 1024 1463.990 30.500 30.440 30.559 23.137 29.720 14.127 2048 2431.283 50.652 50.153 51.155 31.892 50.507 21.675 4096 3730.713 77.723 79.626 75.866 40.586 77.723 29.181 8192 4893.953 101.957 112.274 92.588 65.983 101.957 62.566 16384 5567.697 115.994 136.102 98.856 90.046 115.994 83.488 32768 5906.301 123.048 151.428 99.987 108.601 122.884 97.128 65536 6066.979 126.395 157.909 101.171 118.479 123.992 104.605 131072 6224.382 129.675 162.570 103.435 125.363 123.475 107.663 262144 6341.674 132.118 170.728 102.240 124.517 122.023 106.848 524288 6372.294 132.756 178.745 98.600 121.925 122.481 104.361 1048576 6138.954 127.895 180.245 90.749 114.747 118.795 101.605 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-24*2fix : 0.055 0.871 9.934 63.733 183.911 207.573 -> 74.536 -> 3577.734 MByte/s p01 ring-12*4fix : 0.054 0.848 9.986 86.198 187.129 196.232 -> 77.659 -> 3727.645 MByte/s p02 ring-6*8fix : 0.052 0.825 9.475 80.990 154.503 187.284 -> 69.311 -> 3326.917 MByte/s p03 ring-3*16fix : 0.049 0.779 8.817 82.250 151.119 173.853 -> 66.860 -> 3209.258 MByte/s p04 ring-1*48fix : 0.047 0.752 8.709 83.663 138.655 161.325 -> 63.594 -> 3052.527 MByte/s p05 ring-1*48fix : 0.047 0.751 8.769 83.245 139.158 160.268 -> 63.540 -> 3049.913 MByte/s p06 random-cyc-1dim : 0.044 0.701 8.713 76.987 105.392 103.575 -> 48.936 -> 2348.925 MByte/s p07 random-cyc-1dim : 0.045 0.705 8.481 76.094 97.888 93.111 -> 45.873 -> 2201.911 MByte/s p08 random-cyc-1dim : 0.046 0.734 8.778 77.846 109.711 96.468 -> 50.434 -> 2420.825 MByte/s p09 random-cyc-1dim : 0.045 0.726 8.776 78.384 114.250 109.367 -> 51.525 -> 2473.212 MByte/s p10 random-cyc-1dim : 0.044 0.714 8.848 70.982 83.517 71.474 -> 39.311 -> 1886.918 MByte/s p11 random-cyc-1dim : 0.046 0.719 8.625 75.184 103.073 92.701 -> 47.317 -> 2271.193 MByte/s p12 random-cyc-1dim : 0.045 0.705 8.769 75.504 101.690 89.659 -> 46.557 -> 2234.742 MByte/s p13 random-cyc-1dim : 0.044 0.713 8.764 77.036 107.587 96.201 -> 49.371 -> 2369.787 MByte/s p14 random-cyc-1dim : 0.045 0.735 8.680 78.137 98.935 93.622 -> 46.751 -> 2244.052 MByte/s p15 random-cyc-1dim : 0.046 0.738 8.802 72.854 93.238 69.588 -> 42.364 -> 2033.483 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.042 0.691 8.790 39.178 40.644 38.559 -> 21.598 -> 1036.707 MByte/s p17 best bi-section : 0.034 0.497 5.510 62.938 183.294 209.053 -> 73.355 -> 3521.048 MByte/s p18 worst bi-section : 0.024 0.369 5.325 33.940 38.525 56.004 -> 21.144 -> 1014.908 MByte/s p19 acyclic-1dim-all : 0.047 0.750 8.746 84.786 166.504 139.616 -> 68.114 -> 3269.488 MByte/s p20 acyclic-2dim-all : 0.044 0.772 10.938 80.087 112.081 118.502 -> 53.230 -> 2555.058 MByte/s p21 acyclic-3dim-all : 0.045 0.802 11.003 61.917 77.929 76.579 -> 37.984 -> 1823.229 MByte/s p22 cyclic-1dim-all : 0.047 0.757 8.483 82.202 140.508 161.712 -> 63.169 -> 3032.120 MByte/s p23 cyclic-2dim-all : 0.049 0.847 12.160 80.294 113.551 115.561 -> 52.820 -> 2535.352 MByte/s p24 cyclic-3dim-all : 0.058 1.012 14.202 70.923 95.057 95.803 -> 45.661 -> 2191.715 MByte/s log_avg of all rings : 0.051 0.803 9.266 79.626 157.909 180.245 || 69.050 -> 3314.408 MByte/s log_avg of all random : 0.045 0.719 8.723 75.866 101.171 90.749 || 46.706 -> 2241.874 MByte/s log_avg(ring,random) : 0.048 0.760 8.990 77.723 126.395 127.895 || 56.789 -> 2725.891 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 2725.891 MByte/s on 48 processes ( = 56.789 MByte/s * 48 processes) system parameters : 48 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 = 2725.891 MB/s = 56.789 * 48 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E