b_eff = 2990.339 MB/s = 54.370 * 55 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 = 55 1-dim-paterns: size = 55 2-dim-paterns: size = 11 * 5 3-dim-paterns: size = 6 * 3 * 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 = 156.615 sec sum of max elapsed time per entries above = 158.525 sec difference = -1.910 sec = 1.2% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-27*2&+1 => 1 sendrecv_calls with 55 messages, i.e. msgs/used node, all nodes are used p01 ring-14*4&-1 => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p02 ring-7*8&-1 => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p03 ring-3*18&+1 => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p04 ring-2*28&-1 => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p05 ring-1*55fix => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 54 messages, i.e. msgs/used node, 1 nodes are UNUSED p18 worst bi-section => 2 sendrecv_calls with 54 messages, i.e. msgs/used node, 1 nodes are UNUSED p19 acyclic-1dim-all => 2 sendrecv_calls with 108 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 188 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 234 messages, i.e. msgs/used node, 1 nodes are UNUSED p22 cyclic-1dim-all => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 220 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 6 sendrecv_calls with 324 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-27*2&+1 : 69.058 71.767 57.097 -> 71.767 -> 3947.184 MByte/s p01 ring-14*4&-1 : 60.610 76.300 45.234 -> 76.300 -> 4196.516 MByte/s p02 ring-7*8&-1 : 53.475 68.091 44.012 -> 68.091 -> 3745.015 MByte/s p03 ring-3*18&+1 : 52.793 63.883 44.207 -> 63.883 -> 3513.544 MByte/s p04 ring-2*28&-1 : 54.153 77.214 45.495 -> 77.214 -> 4246.750 MByte/s p05 ring-1*55fix : 52.149 64.528 44.723 -> 64.528 -> 3549.048 MByte/s p06 random-cyc-1dim : 39.008 44.227 34.873 -> 44.227 -> 2432.509 MByte/s p07 random-cyc-1dim : 35.267 34.668 29.955 -> 35.267 -> 1939.705 MByte/s p08 random-cyc-1dim : 35.595 38.698 31.168 -> 38.698 -> 2128.370 MByte/s p09 random-cyc-1dim : 39.442 41.517 33.776 -> 41.517 -> 2283.430 MByte/s p10 random-cyc-1dim : 36.002 34.896 31.409 -> 36.002 -> 1980.087 MByte/s p11 random-cyc-1dim : 40.587 39.867 34.123 -> 40.587 -> 2232.312 MByte/s p12 random-cyc-1dim : 39.931 39.948 35.131 -> 39.948 -> 2197.116 MByte/s p13 random-cyc-1dim : 37.437 38.045 32.238 -> 38.045 -> 2092.496 MByte/s p14 random-cyc-1dim : 33.040 38.539 32.066 -> 38.539 -> 2119.663 MByte/s p15 random-cyc-1dim : 36.616 36.219 30.876 -> 36.616 -> 2013.878 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 20.722 21.818 19.093 -> 21.818 -> 1199.967 MByte/s p17 best bi-section : 50.664 70.385 48.025 -> 70.385 -> 3871.175 MByte/s p18 worst bi-section : 18.678 23.372 18.985 -> 23.372 -> 1285.483 MByte/s p19 acyclic-1dim-all : 53.468 67.249 45.906 -> 67.249 -> 3698.705 MByte/s p20 acyclic-2dim-all : 40.352 47.673 33.824 -> 47.673 -> 2622.019 MByte/s p21 acyclic-3dim-all : 34.505 39.107 29.552 -> 39.107 -> 2150.902 MByte/s p22 cyclic-1dim-all : 52.087 65.421 44.451 -> 65.421 -> 3598.157 MByte/s p23 cyclic-2dim-all : 43.250 47.899 36.194 -> 47.899 -> 2634.430 MByte/s p24 cyclic-3dim-all : 41.099 43.473 34.000 -> 43.473 -> 2391.039 MByte/s log_avg of all rings : 56.740 70.102 46.588 || 70.102 -> 3855.588 MByte/s log_avg of all random : 37.221 38.562 32.517 || 38.860 -> 2137.313 MByte/s log_avg(ring,random) : 45.956 51.993 38.922 ||( 52.194 -> 2870.644)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-27*2&+1 : 72.576 72.857 73.103 -> 73.103 -> 4020.654 MByte/s p01 ring-14*4&-1 : 74.485 76.202 76.313 -> 76.313 -> 4197.232 MByte/s p02 ring-7*8&-1 : 67.571 66.728 66.992 -> 67.571 -> 3716.395 MByte/s p03 ring-3*18&+1 : 62.824 61.985 62.817 -> 62.824 -> 3455.325 MByte/s p04 ring-2*28&-1 : 75.391 75.965 76.708 -> 76.708 -> 4218.945 MByte/s p05 ring-1*55fix : 63.552 63.211 63.557 -> 63.557 -> 3495.624 MByte/s p06 random-cyc-1dim : 44.145 44.335 44.087 -> 44.335 -> 2438.420 MByte/s p07 random-cyc-1dim : 38.080 38.044 38.314 -> 38.314 -> 2107.256 MByte/s p08 random-cyc-1dim : 39.687 39.522 39.654 -> 39.687 -> 2182.769 MByte/s p09 random-cyc-1dim : 43.922 43.225 43.352 -> 43.922 -> 2415.685 MByte/s p10 random-cyc-1dim : 39.871 39.870 38.700 -> 39.871 -> 2192.911 MByte/s p11 random-cyc-1dim : 44.893 45.257 44.873 -> 45.257 -> 2489.112 MByte/s p12 random-cyc-1dim : 44.386 43.934 43.733 -> 44.386 -> 2441.216 MByte/s p13 random-cyc-1dim : 41.104 40.666 41.128 -> 41.128 -> 2262.024 MByte/s p14 random-cyc-1dim : 38.327 38.518 37.877 -> 38.518 -> 2118.517 MByte/s p15 random-cyc-1dim : 39.304 40.331 39.528 -> 40.331 -> 2218.188 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 22.632 22.364 22.509 -> 22.632 -> 1244.756 MByte/s p17 best bi-section : 69.559 70.294 69.884 -> 70.294 -> 3866.161 MByte/s p18 worst bi-section : 23.187 23.203 23.220 -> 23.220 -> 1277.125 MByte/s p19 acyclic-1dim-all : 66.122 66.968 66.734 -> 66.968 -> 3683.227 MByte/s p20 acyclic-2dim-all : 48.748 48.524 48.438 -> 48.748 -> 2681.114 MByte/s p21 acyclic-3dim-all : 42.507 41.783 41.640 -> 42.507 -> 2337.900 MByte/s p22 cyclic-1dim-all : 64.830 63.101 62.814 -> 64.830 -> 3565.655 MByte/s p23 cyclic-2dim-all : 50.158 50.946 49.886 -> 50.946 -> 2802.044 MByte/s p24 cyclic-3dim-all : 48.477 48.057 48.308 -> 48.477 -> 2666.243 MByte/s log_avg of all rings : 69.215 69.248 69.680 || 69.781 -> 3837.955 MByte/s log_avg of all random : 41.293 41.298 41.048 || 41.500 -> 2282.482 MByte/s log_avg(ring,random) : 53.461 53.477 53.481 ||( 53.813 -> 2959.740)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-27*2&+1 p00 method 0 : 0.055 0.869 10.348 51.559 173.846 165.224 -> 69.058 -> 3798.182 MByte/s p00 method 1 : 0.019 0.326 4.977 58.984 181.479 208.438 -> 71.767 -> 3947.184 MByte/s p00 method 2 : 0.024 0.394 4.940 25.992 161.755 129.098 -> 57.097 -> 3140.359 MByte/s p01 ring-14*4&-1 p01 method 0 : 0.054 0.838 9.742 50.605 159.348 153.852 -> 60.610 -> 3333.534 MByte/s p01 method 1 : 0.031 0.546 8.087 82.364 186.050 195.754 -> 76.300 -> 4196.516 MByte/s p01 method 2 : 0.022 0.343 4.626 29.530 119.661 106.170 -> 45.234 -> 2487.884 MByte/s p02 ring-7*8&-1 p02 method 0 : 0.051 0.818 9.520 47.539 132.930 137.707 -> 53.475 -> 2941.152 MByte/s p02 method 1 : 0.031 0.543 8.050 78.253 155.753 186.316 -> 68.091 -> 3745.015 MByte/s p02 method 2 : 0.022 0.339 4.476 29.795 116.570 107.181 -> 44.012 -> 2420.661 MByte/s p03 ring-3*18&+1 p03 method 0 : 0.048 0.781 8.563 40.706 133.858 134.464 -> 52.793 -> 2903.620 MByte/s p03 method 1 : 0.031 0.536 7.924 76.987 150.745 153.531 -> 63.883 -> 3513.544 MByte/s p03 method 2 : 0.022 0.343 4.458 29.896 120.127 105.491 -> 44.207 -> 2431.395 MByte/s p04 ring-2*28&-1 p04 method 0 : 0.049 0.775 8.560 39.644 140.565 132.342 -> 54.153 -> 2978.399 MByte/s p04 method 1 : 0.031 0.541 8.038 80.645 190.972 201.171 -> 77.214 -> 4246.750 MByte/s p04 method 2 : 0.021 0.343 4.431 29.922 123.228 96.169 -> 45.495 -> 2502.213 MByte/s p05 ring-1*55fix p05 method 0 : 0.047 0.759 8.131 38.178 132.515 127.043 -> 52.149 -> 2868.221 MByte/s p05 method 1 : 0.030 0.531 7.878 78.371 150.618 145.463 -> 64.528 -> 3549.048 MByte/s p05 method 2 : 0.021 0.341 4.368 29.899 117.830 121.533 -> 44.723 -> 2459.742 MByte/s p06 random-cyc-1dim p06 method 0 : 0.045 0.728 7.793 37.346 93.503 95.638 -> 39.008 -> 2145.455 MByte/s p06 method 1 : 0.031 0.531 7.978 72.569 95.430 88.723 -> 44.227 -> 2432.509 MByte/s p06 method 2 : 0.021 0.337 4.240 29.325 90.792 89.290 -> 34.873 -> 1918.020 MByte/s p07 random-cyc-1dim p07 method 0 : 0.044 0.707 7.849 37.136 85.525 76.836 -> 35.267 -> 1939.705 MByte/s p07 method 1 : 0.030 0.528 7.940 65.491 76.611 47.748 -> 34.668 -> 1906.742 MByte/s p07 method 2 : 0.021 0.332 4.310 29.188 75.823 57.599 -> 29.955 -> 1647.538 MByte/s p08 random-cyc-1dim p08 method 0 : 0.044 0.705 7.947 37.379 85.022 79.871 -> 35.595 -> 1957.704 MByte/s p08 method 1 : 0.030 0.523 7.820 69.388 85.000 62.775 -> 38.698 -> 2128.370 MByte/s p08 method 2 : 0.021 0.331 4.268 29.087 73.899 71.821 -> 31.168 -> 1714.227 MByte/s p09 random-cyc-1dim p09 method 0 : 0.045 0.725 7.962 37.800 98.818 88.025 -> 39.442 -> 2169.304 MByte/s p09 method 1 : 0.031 0.527 7.870 73.130 97.009 66.573 -> 41.517 -> 2283.430 MByte/s p09 method 2 : 0.021 0.338 4.350 29.250 84.110 72.964 -> 33.776 -> 1857.702 MByte/s p10 random-cyc-1dim p10 method 0 : 0.044 0.711 7.738 37.219 85.777 80.961 -> 36.002 -> 1980.087 MByte/s p10 method 1 : 0.031 0.530 7.939 66.893 79.786 41.407 -> 34.896 -> 1919.271 MByte/s p10 method 2 : 0.021 0.334 4.324 29.340 79.076 70.279 -> 31.409 -> 1727.509 MByte/s p11 random-cyc-1dim p11 method 0 : 0.045 0.726 7.827 37.745 98.347 100.739 -> 40.587 -> 2232.312 MByte/s p11 method 1 : 0.031 0.531 7.942 71.292 88.070 67.742 -> 39.867 -> 2192.700 MByte/s p11 method 2 : 0.021 0.337 4.391 29.355 86.038 76.471 -> 34.123 -> 1876.781 MByte/s p12 random-cyc-1dim p12 method 0 : 0.045 0.722 7.995 37.273 99.350 85.914 -> 39.931 -> 2196.229 MByte/s p12 method 1 : 0.030 0.518 7.798 72.833 87.998 59.519 -> 39.948 -> 2197.116 MByte/s p12 method 2 : 0.021 0.334 4.348 29.457 90.393 80.263 -> 35.131 -> 1932.232 MByte/s p13 random-cyc-1dim p13 method 0 : 0.044 0.713 7.783 37.449 91.441 81.331 -> 37.437 -> 2059.016 MByte/s p13 method 1 : 0.031 0.535 7.988 70.841 84.220 67.275 -> 38.045 -> 2092.496 MByte/s p13 method 2 : 0.021 0.335 4.358 29.337 79.477 79.711 -> 32.238 -> 1773.095 MByte/s p14 random-cyc-1dim p14 method 0 : 0.044 0.697 7.746 36.794 78.196 73.084 -> 33.040 -> 1817.216 MByte/s p14 method 1 : 0.031 0.528 7.810 68.828 83.225 74.878 -> 38.539 -> 2119.663 MByte/s p14 method 2 : 0.021 0.329 4.302 29.039 77.622 75.915 -> 32.066 -> 1763.647 MByte/s p15 random-cyc-1dim p15 method 0 : 0.045 0.694 7.703 36.390 88.634 87.543 -> 36.616 -> 2013.878 MByte/s p15 method 1 : 0.031 0.524 7.851 68.288 80.199 54.725 -> 36.219 -> 1992.034 MByte/s p15 method 2 : 0.021 0.334 4.347 29.204 76.020 76.121 -> 30.876 -> 1698.154 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.043 0.677 7.612 32.786 44.600 32.313 -> 20.722 -> 1139.729 MByte/s p16 method 1 : 0.031 0.531 7.950 42.414 42.582 32.000 -> 21.818 -> 1199.967 MByte/s p16 method 2 : 0.021 0.329 4.193 27.656 42.138 38.358 -> 19.093 -> 1050.130 MByte/s p17 best bi-section p17 method 0 : 0.032 0.489 5.458 27.586 128.177 160.348 -> 50.664 -> 2786.539 MByte/s p17 method 1 : 0.019 0.314 4.794 56.991 177.981 204.653 -> 70.385 -> 3871.175 MByte/s p17 method 2 : 0.015 0.234 3.138 24.006 122.532 158.389 -> 48.025 -> 2641.367 MByte/s p18 worst bi-section p18 method 0 : 0.023 0.362 4.615 25.518 42.339 34.005 -> 18.678 -> 1027.287 MByte/s p18 method 1 : 0.018 0.307 4.690 37.330 46.371 56.293 -> 23.372 -> 1285.483 MByte/s p18 method 2 : 0.014 0.231 3.260 22.765 43.875 43.285 -> 18.985 -> 1044.191 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.047 0.748 8.261 38.423 140.935 128.763 -> 53.468 -> 2940.752 MByte/s p19 method 1 : 0.031 0.536 7.954 80.580 166.475 143.619 -> 67.249 -> 3698.705 MByte/s p19 method 2 : 0.021 0.335 4.397 29.479 122.081 118.639 -> 45.906 -> 2524.847 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.035 0.548 5.968 28.184 104.377 105.624 -> 40.352 -> 2219.365 MByte/s p20 method 1 : 0.042 0.725 10.291 76.449 102.671 100.706 -> 47.673 -> 2622.019 MByte/s p20 method 2 : 0.018 0.291 3.739 25.408 89.876 70.296 -> 33.824 -> 1860.311 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.030 0.474 5.313 25.608 86.276 86.567 -> 34.505 -> 1897.787 MByte/s p21 method 1 : 0.041 0.738 10.343 71.792 81.281 67.945 -> 39.107 -> 2150.902 MByte/s p21 method 2 : 0.015 0.240 3.075 21.093 81.099 59.271 -> 29.552 -> 1625.377 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.047 0.752 8.166 38.143 135.489 131.943 -> 52.087 -> 2864.791 MByte/s p22 method 1 : 0.031 0.520 7.812 78.835 159.067 166.593 -> 65.421 -> 3598.157 MByte/s p22 method 2 : 0.022 0.342 4.501 29.890 115.016 105.928 -> 44.451 -> 2444.823 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.039 0.617 6.814 31.873 108.514 110.265 -> 43.250 -> 2378.775 MByte/s p23 method 1 : 0.045 0.786 11.354 78.213 103.736 96.245 -> 47.899 -> 2634.430 MByte/s p23 method 2 : 0.021 0.339 4.377 29.668 93.872 75.783 -> 36.194 -> 1990.667 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.037 0.583 6.426 30.853 103.519 102.628 -> 41.099 -> 2260.419 MByte/s p24 method 1 : 0.054 0.953 13.384 74.679 88.364 76.469 -> 43.473 -> 2391.039 MByte/s p24 method 2 : 0.021 0.328 4.197 28.670 87.730 74.583 -> 34.000 -> 1870.015 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.806 9.111 44.381 144.706 141.166 || 56.740 -> 3120.725 MByte/s - ring, method 1 : 0.029 0.496 7.388 75.484 168.396 180.126 || 70.102 -> 3855.588 MByte/s - ring, method 2 : 0.022 0.350 4.546 29.135 125.652 110.407 || 46.588 -> 2562.332 MByte/s log_avg of all random - random, method 0 : 0.045 0.713 7.834 37.251 90.209 84.626 || 37.221 -> 2047.170 MByte/s - random, method 1 : 0.031 0.527 7.893 69.911 85.532 61.817 || 38.562 -> 2120.888 MByte/s - random, method 2 : 0.021 0.334 4.324 29.258 81.121 74.624 || 32.517 -> 1788.441 MByte/s log_avg(ring,random) - average, method 0 : 0.047 0.758 8.448 40.660 114.253 109.299 || 45.956 -> 2527.579 MByte/s - average, method 1 : 0.030 0.511 7.636 72.644 120.013 105.521 || 51.993 -> 2859.593 MByte/s - average, method 2 : 0.022 0.342 4.433 29.197 100.961 90.769 || 38.922 -> 2140.696 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 2.612 0.047 0.051 0.045 0.047 0.030 0.022 2 5.200 0.095 0.101 0.089 0.095 0.059 0.043 4 10.485 0.191 0.204 0.179 0.191 0.130 0.086 8 20.888 0.380 0.405 0.356 0.380 0.260 0.172 16 41.688 0.758 0.806 0.713 0.758 0.511 0.342 32 63.223 1.150 1.261 1.048 1.148 1.003 0.596 64 117.539 2.137 2.290 1.994 2.103 1.931 1.138 128 253.007 4.600 5.011 4.223 4.599 3.991 2.355 256 467.645 8.503 9.111 7.935 8.448 7.636 4.433 512 860.463 15.645 16.290 15.026 15.007 14.581 8.118 1024 1550.056 28.183 28.552 27.819 23.212 27.236 13.961 2048 2597.980 47.236 47.611 46.864 31.621 46.869 21.549 4096 3995.420 72.644 75.484 69.911 40.660 72.644 29.197 8192 5224.067 94.983 108.501 83.149 65.347 94.983 62.413 16384 5973.196 108.604 136.159 86.624 88.182 108.604 81.750 32768 6490.517 118.009 158.357 87.942 104.991 117.195 94.347 65536 6806.911 123.762 168.396 90.959 114.253 120.013 100.961 131072 6959.361 126.534 174.391 91.810 117.823 119.328 101.897 262144 6998.828 127.251 179.162 90.382 117.920 117.453 102.500 524288 7017.325 127.588 184.574 88.196 115.964 113.642 97.653 1048576 6803.421 123.699 180.126 84.948 109.299 105.521 90.769 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-27*2&+1 : 0.055 0.869 10.348 58.984 181.479 208.438 -> 73.330 -> 4033.169 MByte/s p01 ring-14*4&-1 : 0.054 0.838 9.742 82.364 186.050 195.754 -> 76.600 -> 4212.989 MByte/s p02 ring-7*8&-1 : 0.051 0.818 9.520 78.253 155.753 186.316 -> 68.337 -> 3758.522 MByte/s p03 ring-3*18&+1 : 0.048 0.781 8.563 76.987 150.745 153.531 -> 63.984 -> 3519.106 MByte/s p04 ring-2*28&-1 : 0.049 0.775 8.560 80.645 190.972 201.171 -> 77.295 -> 4251.201 MByte/s p05 ring-1*55fix : 0.047 0.759 8.131 78.371 150.618 145.463 -> 64.584 -> 3552.129 MByte/s p06 random-cyc-1dim : 0.045 0.728 7.978 72.569 95.430 95.638 -> 45.034 -> 2476.863 MByte/s p07 random-cyc-1dim : 0.044 0.707 7.940 65.491 85.525 76.836 -> 39.049 -> 2147.668 MByte/s p08 random-cyc-1dim : 0.044 0.705 7.947 69.388 85.022 79.871 -> 40.006 -> 2200.321 MByte/s p09 random-cyc-1dim : 0.045 0.725 7.962 73.130 98.818 88.025 -> 44.404 -> 2442.198 MByte/s p10 random-cyc-1dim : 0.044 0.711 7.939 66.893 85.777 80.961 -> 40.189 -> 2210.413 MByte/s p11 random-cyc-1dim : 0.045 0.726 7.942 71.292 98.347 100.739 -> 45.673 -> 2512.031 MByte/s p12 random-cyc-1dim : 0.045 0.722 7.995 72.833 99.350 85.914 -> 44.860 -> 2467.310 MByte/s p13 random-cyc-1dim : 0.044 0.713 7.988 70.841 91.441 81.331 -> 41.790 -> 2298.462 MByte/s p14 random-cyc-1dim : 0.044 0.697 7.810 68.828 83.225 75.915 -> 38.657 -> 2126.147 MByte/s p15 random-cyc-1dim : 0.045 0.694 7.851 68.288 88.634 87.543 -> 40.521 -> 2228.648 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.043 0.677 7.950 42.414 44.600 38.358 -> 22.821 -> 1255.178 MByte/s p17 best bi-section : 0.032 0.489 5.458 56.991 177.981 204.653 -> 70.488 -> 3876.841 MByte/s p18 worst bi-section : 0.023 0.362 4.690 37.330 46.371 56.293 -> 23.385 -> 1286.184 MByte/s p19 acyclic-1dim-all : 0.047 0.748 8.261 80.580 166.475 143.619 -> 67.308 -> 3701.966 MByte/s p20 acyclic-2dim-all : 0.042 0.725 10.291 76.449 104.377 105.624 -> 49.181 -> 2704.956 MByte/s p21 acyclic-3dim-all : 0.041 0.738 10.343 71.792 86.276 86.567 -> 42.607 -> 2343.382 MByte/s p22 cyclic-1dim-all : 0.047 0.752 8.166 78.835 159.067 166.593 -> 65.486 -> 3601.712 MByte/s p23 cyclic-2dim-all : 0.045 0.786 11.354 78.213 108.514 110.265 -> 51.107 -> 2810.882 MByte/s p24 cyclic-3dim-all : 0.054 0.953 13.384 74.679 103.519 102.628 -> 48.901 -> 2689.544 MByte/s log_avg of all rings : 0.051 0.806 9.111 75.484 168.396 180.126 || 70.483 -> 3876.571 MByte/s log_avg of all random : 0.045 0.713 7.935 69.911 90.959 84.948 || 41.940 -> 2306.710 MByte/s log_avg(ring,random) : 0.047 0.758 8.503 72.644 123.762 123.699 || 54.370 -> 2990.339 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 2990.339 MByte/s on 55 processes ( = 54.370 MByte/s * 55 processes) system parameters : 55 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 = 2990.339 MB/s = 54.370 * 55 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E