b_eff = 2949.705 MB/s = 56.725 * 52 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 = 52 1-dim-paterns: size = 52 2-dim-paterns: size = 13 * 4 3-dim-paterns: size = 13 * 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 = 142.229 sec sum of max elapsed time per entries above = 145.360 sec difference = -3.131 sec = 2.2% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-26*2fix => 1 sendrecv_calls with 52 messages, i.e. msgs/used node, all nodes are used p01 ring-13*4fix => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p02 ring-6*8&+1 => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p03 ring-3*18&-1 => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p04 ring-2*26fix => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p05 ring-1*52fix => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 52 messages, i.e. msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 52 messages, i.e. msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 102 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 174 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 200 messages, i.e. msgs/used node, all nodes are used p22 cyclic-1dim-all => 2 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 208 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 4 sendrecv_calls with 208 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-26*2fix : 71.934 72.543 61.074 -> 72.543 -> 3772.256 MByte/s p01 ring-13*4fix : 59.550 76.451 45.518 -> 76.451 -> 3975.460 MByte/s p02 ring-6*8&+1 : 53.315 66.742 45.955 -> 66.742 -> 3470.563 MByte/s p03 ring-3*18&-1 : 53.778 71.144 45.850 -> 71.144 -> 3699.479 MByte/s p04 ring-2*26fix : 54.876 79.321 46.468 -> 79.321 -> 4124.689 MByte/s p05 ring-1*52fix : 53.878 76.138 46.590 -> 76.138 -> 3959.185 MByte/s p06 random-cyc-1dim : 37.522 41.038 33.678 -> 41.038 -> 2133.976 MByte/s p07 random-cyc-1dim : 38.859 42.019 33.772 -> 42.019 -> 2184.965 MByte/s p08 random-cyc-1dim : 38.186 38.283 33.253 -> 38.283 -> 1990.701 MByte/s p09 random-cyc-1dim : 38.451 40.454 32.393 -> 40.454 -> 2103.620 MByte/s p10 random-cyc-1dim : 43.112 41.331 36.805 -> 43.112 -> 2241.848 MByte/s p11 random-cyc-1dim : 36.184 38.756 32.835 -> 38.756 -> 2015.326 MByte/s p12 random-cyc-1dim : 42.063 47.261 36.206 -> 47.261 -> 2457.598 MByte/s p13 random-cyc-1dim : 36.418 40.554 32.510 -> 40.554 -> 2108.823 MByte/s p14 random-cyc-1dim : 36.136 38.424 32.182 -> 38.424 -> 1998.032 MByte/s p15 random-cyc-1dim : 35.930 34.599 30.671 -> 35.930 -> 1868.356 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 23.602 24.687 21.693 -> 24.687 -> 1283.701 MByte/s p17 best bi-section : 51.565 72.443 49.373 -> 72.443 -> 3767.029 MByte/s p18 worst bi-section : 18.890 23.795 18.910 -> 23.795 -> 1237.334 MByte/s p19 acyclic-1dim-all : 53.947 67.456 46.598 -> 67.456 -> 3507.694 MByte/s p20 acyclic-2dim-all : 42.852 48.623 34.994 -> 48.623 -> 2528.419 MByte/s p21 acyclic-3dim-all : 42.169 49.659 37.527 -> 49.659 -> 2582.268 MByte/s p22 cyclic-1dim-all : 54.368 75.551 47.225 -> 75.551 -> 3928.630 MByte/s p23 cyclic-2dim-all : 46.520 49.204 38.157 -> 49.204 -> 2558.603 MByte/s p24 cyclic-3dim-all : 46.326 53.466 39.831 -> 53.466 -> 2780.222 MByte/s log_avg of all rings : 57.545 73.607 48.290 || 73.607 -> 3827.551 MByte/s log_avg of all random : 38.215 40.155 33.385 || 40.478 -> 2104.831 MByte/s log_avg(ring,random) : 46.894 54.366 40.152 ||( 54.584 -> 2838.371)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-26*2fix : 73.561 73.194 73.752 -> 73.752 -> 3835.109 MByte/s p01 ring-13*4fix : 75.309 76.098 74.933 -> 76.098 -> 3957.101 MByte/s p02 ring-6*8&+1 : 65.670 65.501 65.999 -> 65.999 -> 3431.934 MByte/s p03 ring-3*18&-1 : 69.741 70.433 70.432 -> 70.433 -> 3662.520 MByte/s p04 ring-2*26fix : 78.318 78.063 77.572 -> 78.318 -> 4072.543 MByte/s p05 ring-1*52fix : 74.188 75.537 74.554 -> 75.537 -> 3927.944 MByte/s p06 random-cyc-1dim : 41.984 42.165 41.781 -> 42.165 -> 2192.581 MByte/s p07 random-cyc-1dim : 43.718 43.863 43.221 -> 43.863 -> 2280.889 MByte/s p08 random-cyc-1dim : 42.827 42.899 42.607 -> 42.899 -> 2230.738 MByte/s p09 random-cyc-1dim : 42.969 43.403 43.630 -> 43.630 -> 2268.740 MByte/s p10 random-cyc-1dim : 48.373 48.068 48.447 -> 48.447 -> 2519.253 MByte/s p11 random-cyc-1dim : 41.157 41.145 41.699 -> 41.699 -> 2168.356 MByte/s p12 random-cyc-1dim : 47.651 46.904 47.644 -> 47.651 -> 2477.861 MByte/s p13 random-cyc-1dim : 40.555 41.631 40.538 -> 41.631 -> 2164.790 MByte/s p14 random-cyc-1dim : 39.830 39.718 39.298 -> 39.830 -> 2071.165 MByte/s p15 random-cyc-1dim : 38.822 39.004 38.340 -> 39.004 -> 2028.233 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 25.489 25.675 25.392 -> 25.675 -> 1335.114 MByte/s p17 best bi-section : 72.075 72.286 71.410 -> 72.286 -> 3758.898 MByte/s p18 worst bi-section : 23.664 23.582 23.741 -> 23.741 -> 1234.538 MByte/s p19 acyclic-1dim-all : 65.775 65.961 66.119 -> 66.119 -> 3438.197 MByte/s p20 acyclic-2dim-all : 51.228 51.547 51.783 -> 51.783 -> 2692.701 MByte/s p21 acyclic-3dim-all : 52.519 52.362 52.353 -> 52.519 -> 2731.008 MByte/s p22 cyclic-1dim-all : 74.740 74.876 75.058 -> 75.058 -> 3903.038 MByte/s p23 cyclic-2dim-all : 53.581 54.691 54.176 -> 54.691 -> 2843.933 MByte/s p24 cyclic-3dim-all : 54.671 55.021 54.612 -> 55.021 -> 2861.068 MByte/s log_avg of all rings : 72.682 73.015 72.776 || 73.240 -> 3808.472 MByte/s log_avg of all random : 42.688 42.795 42.611 || 42.988 -> 2235.384 MByte/s log_avg(ring,random) : 55.701 55.899 55.687 ||( 56.111 -> 2917.773)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-26*2fix p00 method 0 : 0.055 0.871 10.047 51.304 179.633 208.305 -> 71.934 -> 3740.562 MByte/s p00 method 1 : 0.020 0.342 5.243 61.171 183.830 208.049 -> 72.543 -> 3772.256 MByte/s p00 method 2 : 0.025 0.392 5.286 26.305 169.162 163.063 -> 61.074 -> 3175.848 MByte/s p01 ring-13*4fix p01 method 0 : 0.054 0.853 10.007 50.010 148.313 152.851 -> 59.550 -> 3096.598 MByte/s p01 method 1 : 0.033 0.571 8.581 84.518 186.139 194.430 -> 76.451 -> 3975.460 MByte/s p01 method 2 : 0.022 0.364 4.585 29.554 121.938 107.904 -> 45.518 -> 2366.926 MByte/s p02 ring-6*8&+1 p02 method 0 : 0.052 0.824 9.432 47.991 133.776 132.195 -> 53.315 -> 2772.406 MByte/s p02 method 1 : 0.033 0.570 8.510 83.507 150.194 180.636 -> 66.742 -> 3470.563 MByte/s p02 method 2 : 0.022 0.344 4.402 29.984 121.552 126.407 -> 45.955 -> 2389.657 MByte/s p03 ring-3*18&-1 p03 method 0 : 0.049 0.797 9.151 45.641 136.614 132.139 -> 53.778 -> 2796.463 MByte/s p03 method 1 : 0.033 0.569 8.467 80.676 160.445 194.591 -> 71.144 -> 3699.479 MByte/s p03 method 2 : 0.022 0.345 4.433 29.989 120.734 118.027 -> 45.850 -> 2384.221 MByte/s p04 ring-2*26fix p04 method 0 : 0.048 0.790 8.611 43.947 142.303 133.288 -> 54.876 -> 2853.553 MByte/s p04 method 1 : 0.033 0.573 8.501 83.386 182.493 212.174 -> 79.321 -> 4124.689 MByte/s p04 method 2 : 0.022 0.344 4.470 30.174 125.656 113.041 -> 46.468 -> 2416.351 MByte/s p05 ring-1*52fix p05 method 0 : 0.048 0.767 8.337 38.527 138.520 134.144 -> 53.878 -> 2801.641 MByte/s p05 method 1 : 0.032 0.562 8.405 84.900 181.008 201.521 -> 76.138 -> 3959.185 MByte/s p05 method 2 : 0.022 0.342 4.462 29.562 123.230 118.985 -> 46.590 -> 2422.698 MByte/s p06 random-cyc-1dim p06 method 0 : 0.045 0.709 7.965 37.556 93.099 80.657 -> 37.522 -> 1951.150 MByte/s p06 method 1 : 0.032 0.558 8.223 72.123 86.764 73.264 -> 41.038 -> 2133.976 MByte/s p06 method 2 : 0.021 0.335 4.184 29.485 86.941 77.094 -> 33.678 -> 1751.269 MByte/s p07 random-cyc-1dim p07 method 0 : 0.045 0.708 7.640 36.519 97.103 87.121 -> 38.859 -> 2020.659 MByte/s p07 method 1 : 0.032 0.558 8.198 72.132 93.902 70.742 -> 42.019 -> 2184.965 MByte/s p07 method 2 : 0.021 0.333 4.254 29.487 87.870 71.099 -> 33.772 -> 1756.146 MByte/s p08 random-cyc-1dim p08 method 0 : 0.044 0.696 7.780 37.671 96.447 80.469 -> 38.186 -> 1985.658 MByte/s p08 method 1 : 0.032 0.550 8.225 71.751 83.711 51.401 -> 38.283 -> 1990.701 MByte/s p08 method 2 : 0.021 0.336 4.312 29.207 81.300 75.409 -> 33.253 -> 1729.155 MByte/s p09 random-cyc-1dim p09 method 0 : 0.044 0.725 7.852 38.150 94.481 79.897 -> 38.451 -> 1999.457 MByte/s p09 method 1 : 0.033 0.557 8.293 74.015 89.773 62.353 -> 40.454 -> 2103.620 MByte/s p09 method 2 : 0.020 0.334 4.373 29.488 82.181 65.062 -> 32.393 -> 1684.422 MByte/s p10 random-cyc-1dim p10 method 0 : 0.041 0.728 7.911 38.132 106.810 99.352 -> 43.112 -> 2241.848 MByte/s p10 method 1 : 0.032 0.557 8.357 76.086 90.805 53.650 -> 41.331 -> 2149.228 MByte/s p10 method 2 : 0.021 0.337 4.386 29.390 92.892 83.076 -> 36.805 -> 1913.853 MByte/s p11 random-cyc-1dim p11 method 0 : 0.041 0.706 7.704 37.899 90.615 65.683 -> 36.184 -> 1881.573 MByte/s p11 method 1 : 0.032 0.554 8.304 71.878 85.526 51.341 -> 38.756 -> 2015.326 MByte/s p11 method 2 : 0.021 0.333 4.351 29.090 80.510 77.764 -> 32.835 -> 1707.431 MByte/s p12 random-cyc-1dim p12 method 0 : 0.044 0.719 7.560 37.713 104.125 104.115 -> 42.063 -> 2187.261 MByte/s p12 method 1 : 0.033 0.569 8.491 74.889 104.531 94.821 -> 47.261 -> 2457.598 MByte/s p12 method 2 : 0.021 0.334 4.272 29.431 91.648 88.163 -> 36.206 -> 1882.723 MByte/s p13 random-cyc-1dim p13 method 0 : 0.045 0.717 7.875 37.288 90.383 75.602 -> 36.418 -> 1893.756 MByte/s p13 method 1 : 0.032 0.553 8.249 67.638 85.978 71.575 -> 40.554 -> 2108.823 MByte/s p13 method 2 : 0.021 0.333 4.365 29.214 78.694 85.886 -> 32.510 -> 1690.500 MByte/s p14 random-cyc-1dim p14 method 0 : 0.045 0.730 7.744 37.371 87.437 82.170 -> 36.136 -> 1879.093 MByte/s p14 method 1 : 0.033 0.565 8.441 69.645 79.539 76.510 -> 38.424 -> 1998.032 MByte/s p14 method 2 : 0.021 0.337 4.311 29.195 80.903 71.956 -> 32.182 -> 1673.441 MByte/s p15 random-cyc-1dim p15 method 0 : 0.044 0.724 7.700 37.546 83.579 90.870 -> 35.930 -> 1868.356 MByte/s p15 method 1 : 0.032 0.554 8.271 66.224 75.046 53.332 -> 34.599 -> 1799.171 MByte/s p15 method 2 : 0.021 0.331 4.352 29.272 74.150 70.705 -> 30.671 -> 1594.882 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.043 0.717 7.792 34.011 49.848 43.835 -> 23.602 -> 1227.295 MByte/s p16 method 1 : 0.033 0.555 8.415 46.321 49.702 37.371 -> 24.687 -> 1283.701 MByte/s p16 method 2 : 0.021 0.332 4.287 28.578 47.549 48.975 -> 21.693 -> 1128.043 MByte/s p17 best bi-section p17 method 0 : 0.033 0.500 5.511 28.051 130.338 163.036 -> 51.565 -> 2681.395 MByte/s p17 method 1 : 0.020 0.340 5.169 60.580 182.238 208.656 -> 72.443 -> 3767.029 MByte/s p17 method 2 : 0.015 0.236 3.379 27.455 125.124 161.708 -> 49.373 -> 2567.397 MByte/s p18 worst bi-section p18 method 0 : 0.024 0.372 4.736 25.986 42.561 35.944 -> 18.890 -> 982.280 MByte/s p18 method 1 : 0.020 0.331 5.065 37.811 47.305 56.299 -> 23.795 -> 1237.334 MByte/s p18 method 2 : 0.015 0.235 3.379 25.455 43.474 38.562 -> 18.910 -> 983.306 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.048 0.772 8.362 38.381 140.253 128.865 -> 53.947 -> 2805.226 MByte/s p19 method 1 : 0.033 0.560 8.369 82.298 167.456 141.013 -> 67.456 -> 3507.694 MByte/s p19 method 2 : 0.021 0.335 4.363 29.653 122.500 127.773 -> 46.598 -> 2423.101 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.035 0.550 6.029 27.377 111.734 115.149 -> 42.852 -> 2228.294 MByte/s p20 method 1 : 0.042 0.735 10.390 78.601 108.112 93.484 -> 48.623 -> 2528.419 MByte/s p20 method 2 : 0.018 0.287 3.706 25.515 93.000 76.757 -> 34.994 -> 1819.680 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.031 0.498 5.665 27.608 107.462 115.212 -> 42.169 -> 2192.792 MByte/s p21 method 1 : 0.048 0.844 11.868 81.377 109.588 91.621 -> 49.659 -> 2582.268 MByte/s p21 method 2 : 0.017 0.267 3.541 26.224 102.642 81.920 -> 37.527 -> 1951.426 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.048 0.763 8.478 40.488 142.986 132.699 -> 54.368 -> 2827.141 MByte/s p22 method 1 : 0.031 0.545 8.208 83.258 185.276 197.610 -> 75.551 -> 3928.630 MByte/s p22 method 2 : 0.021 0.346 4.506 30.034 125.390 130.626 -> 47.225 -> 2455.691 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.040 0.640 7.029 33.035 120.223 113.545 -> 46.520 -> 2419.043 MByte/s p23 method 1 : 0.047 0.821 11.815 82.219 107.803 84.965 -> 49.204 -> 2558.603 MByte/s p23 method 2 : 0.021 0.342 4.414 29.759 100.037 85.609 -> 38.157 -> 1984.156 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.039 0.623 6.795 31.463 118.998 119.910 -> 46.326 -> 2408.951 MByte/s p24 method 1 : 0.047 0.826 11.826 81.911 112.588 117.690 -> 53.466 -> 2780.222 MByte/s p24 method 2 : 0.021 0.342 4.394 29.506 105.136 87.575 -> 39.831 -> 2071.234 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.816 9.242 46.033 145.779 146.643 || 57.545 -> 2992.364 MByte/s - ring, method 1 : 0.030 0.523 7.836 79.188 173.459 198.295 || 73.607 -> 3827.551 MByte/s - ring, method 2 : 0.022 0.355 4.597 29.229 129.367 123.399 || 48.290 -> 2511.101 MByte/s log_avg of all random - random, method 0 : 0.044 0.716 7.772 37.582 94.168 83.921 || 38.215 -> 1987.174 MByte/s - random, method 1 : 0.032 0.557 8.305 71.577 87.231 64.607 || 40.155 -> 2088.075 MByte/s - random, method 2 : 0.021 0.334 4.316 29.326 83.520 76.307 || 33.385 -> 1736.037 MByte/s log_avg(ring,random) - average, method 0 : 0.047 0.765 8.475 41.593 117.165 110.935 || 46.894 -> 2438.514 MByte/s - average, method 1 : 0.031 0.540 8.067 75.287 123.008 113.187 || 54.366 -> 2827.050 MByte/s - average, method 2 : 0.022 0.344 4.454 29.277 103.946 97.037 || 40.152 -> 2087.909 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 2.450 0.047 0.051 0.044 0.047 0.031 0.022 2 4.961 0.095 0.102 0.089 0.095 0.063 0.043 4 9.839 0.189 0.202 0.177 0.189 0.138 0.086 8 19.746 0.380 0.408 0.353 0.380 0.275 0.173 16 39.757 0.765 0.816 0.716 0.765 0.540 0.344 32 61.420 1.181 1.274 1.095 1.156 1.058 0.596 64 115.232 2.216 2.335 2.103 2.120 2.037 1.136 128 245.455 4.720 5.112 4.358 4.636 4.226 2.356 256 455.859 8.767 9.254 8.305 8.475 8.067 4.454 512 841.907 16.191 16.619 15.773 15.143 15.290 8.118 1024 1530.151 29.426 29.755 29.100 23.222 28.509 14.070 2048 2564.218 49.312 49.385 49.239 31.709 49.068 21.562 4096 3914.912 75.287 79.188 71.577 41.593 75.287 29.277 8192 5121.687 98.494 112.449 86.270 65.949 98.494 62.100 16384 5861.441 112.720 140.562 90.393 88.909 112.720 82.983 32768 6371.046 122.520 162.537 92.356 106.250 121.697 96.725 65536 6647.198 127.831 173.459 94.205 117.165 123.008 103.946 131072 6897.673 132.648 183.991 95.632 123.559 125.625 106.788 262144 6952.786 133.707 189.437 94.373 121.382 122.946 105.242 524288 6902.462 132.740 194.845 90.430 117.778 119.537 102.996 1048576 6808.882 130.940 198.335 86.446 110.935 113.187 97.037 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-26*2fix : 0.055 0.871 10.047 61.171 183.830 208.305 -> 73.999 -> 3847.961 MByte/s p01 ring-13*4fix : 0.054 0.853 10.007 84.518 186.139 194.430 -> 76.668 -> 3986.759 MByte/s p02 ring-6*8&+1 : 0.052 0.824 9.432 83.507 150.194 180.636 -> 66.887 -> 3478.145 MByte/s p03 ring-3*18&-1 : 0.049 0.797 9.151 80.676 160.445 194.591 -> 71.234 -> 3704.158 MByte/s p04 ring-2*26fix : 0.048 0.790 8.611 83.386 182.493 212.174 -> 79.376 -> 4127.570 MByte/s p05 ring-1*52fix : 0.048 0.767 8.405 84.900 181.008 201.521 -> 76.167 -> 3960.684 MByte/s p06 random-cyc-1dim : 0.045 0.709 8.223 72.123 93.099 80.657 -> 42.608 -> 2215.605 MByte/s p07 random-cyc-1dim : 0.045 0.708 8.198 72.132 97.103 87.121 -> 44.426 -> 2310.133 MByte/s p08 random-cyc-1dim : 0.044 0.696 8.225 71.751 96.447 80.469 -> 43.222 -> 2247.557 MByte/s p09 random-cyc-1dim : 0.044 0.725 8.293 74.015 94.481 79.897 -> 43.962 -> 2286.014 MByte/s p10 random-cyc-1dim : 0.041 0.728 8.357 76.086 106.810 99.352 -> 49.270 -> 2562.041 MByte/s p11 random-cyc-1dim : 0.041 0.706 8.304 71.878 90.615 77.764 -> 42.187 -> 2193.715 MByte/s p12 random-cyc-1dim : 0.044 0.719 8.491 74.889 104.531 104.115 -> 48.203 -> 2506.538 MByte/s p13 random-cyc-1dim : 0.045 0.717 8.249 67.638 90.383 85.886 -> 41.996 -> 2183.805 MByte/s p14 random-cyc-1dim : 0.045 0.730 8.441 69.645 87.437 82.170 -> 40.446 -> 2103.199 MByte/s p15 random-cyc-1dim : 0.044 0.724 8.271 66.224 83.579 90.870 -> 39.792 -> 2069.182 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.043 0.717 8.415 46.321 49.848 48.975 -> 25.889 -> 1346.253 MByte/s p17 best bi-section : 0.033 0.500 5.511 60.580 182.238 208.656 -> 72.519 -> 3770.983 MByte/s p18 worst bi-section : 0.024 0.372 5.065 37.811 47.305 56.299 -> 23.802 -> 1237.726 MByte/s p19 acyclic-1dim-all : 0.048 0.772 8.369 82.298 167.456 141.013 -> 67.497 -> 3509.851 MByte/s p20 acyclic-2dim-all : 0.042 0.735 10.390 78.601 111.734 115.149 -> 52.129 -> 2710.724 MByte/s p21 acyclic-3dim-all : 0.048 0.844 11.868 81.377 109.588 115.212 -> 52.844 -> 2747.884 MByte/s p22 cyclic-1dim-all : 0.048 0.763 8.478 83.258 185.276 197.610 -> 75.603 -> 3931.362 MByte/s p23 cyclic-2dim-all : 0.047 0.821 11.815 82.219 120.223 113.545 -> 54.869 -> 2853.174 MByte/s p24 cyclic-3dim-all : 0.047 0.826 11.826 81.911 118.998 119.910 -> 55.276 -> 2874.372 MByte/s log_avg of all rings : 0.051 0.816 9.254 79.188 173.459 198.335 || 73.942 -> 3844.962 MByte/s log_avg of all random : 0.044 0.716 8.305 71.577 94.205 86.446 || 43.517 -> 2262.899 MByte/s log_avg(ring,random) : 0.047 0.765 8.767 75.287 127.831 130.940 || 56.725 -> 2949.705 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 2949.705 MByte/s on 52 processes ( = 56.725 MByte/s * 52 processes) system parameters : 52 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 = 2949.705 MB/s = 56.725 * 52 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E