b_eff = 3064.914 MB/s = 53.770 * 57 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 = 57 1-dim-paterns: size = 57 2-dim-paterns: size = 19 * 3 3-dim-paterns: size = 7 * 4 * 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 = 151.517 sec sum of max elapsed time per entries above = 152.916 sec difference = -1.399 sec = 0.9% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-28*2&+1 => 1 sendrecv_calls with 57 messages, i.e. msgs/used node, all nodes are used p01 ring-14*4&+1 => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p02 ring-7*8&+1 => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p03 ring-3*19fix => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p04 ring-2*28&+1 => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p05 ring-1*57fix => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 56 messages, i.e. msgs/used node, 1 nodes are UNUSED p18 worst bi-section => 2 sendrecv_calls with 56 messages, i.e. msgs/used node, 1 nodes are UNUSED p19 acyclic-1dim-all => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 184 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 236 messages, i.e. msgs/used node, 1 nodes are UNUSED p22 cyclic-1dim-all => 2 sendrecv_calls with 114 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 228 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 5 sendrecv_calls with 280 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-28*2&+1 : 67.587 71.170 49.119 -> 71.170 -> 4056.693 MByte/s p01 ring-14*4&+1 : 58.405 75.838 44.588 -> 75.838 -> 4322.783 MByte/s p02 ring-7*8&+1 : 53.611 68.427 44.803 -> 68.427 -> 3900.348 MByte/s p03 ring-3*19fix : 52.789 65.671 43.533 -> 65.671 -> 3743.237 MByte/s p04 ring-2*28&+1 : 55.035 77.818 45.648 -> 77.818 -> 4435.605 MByte/s p05 ring-1*57fix : 54.241 77.265 45.451 -> 77.265 -> 4404.093 MByte/s p06 random-cyc-1dim : 37.360 40.548 33.213 -> 40.548 -> 2311.255 MByte/s p07 random-cyc-1dim : 40.058 45.149 35.377 -> 45.149 -> 2573.496 MByte/s p08 random-cyc-1dim : 35.121 37.206 30.905 -> 37.206 -> 2120.762 MByte/s p09 random-cyc-1dim : 37.568 37.542 31.947 -> 37.568 -> 2141.392 MByte/s p10 random-cyc-1dim : 35.266 37.238 30.039 -> 37.238 -> 2122.593 MByte/s p11 random-cyc-1dim : 36.951 41.268 33.756 -> 41.268 -> 2352.257 MByte/s p12 random-cyc-1dim : 33.132 36.958 30.106 -> 36.958 -> 2106.617 MByte/s p13 random-cyc-1dim : 33.760 32.444 28.457 -> 33.760 -> 1924.297 MByte/s p14 random-cyc-1dim : 34.652 35.028 30.277 -> 35.028 -> 1996.615 MByte/s p15 random-cyc-1dim : 30.706 33.845 28.688 -> 33.845 -> 1929.160 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 24.248 24.794 22.382 -> 24.794 -> 1413.277 MByte/s p17 best bi-section : 50.524 69.900 47.717 -> 69.900 -> 3984.276 MByte/s p18 worst bi-section : 17.800 20.685 17.238 -> 20.685 -> 1179.040 MByte/s p19 acyclic-1dim-all : 54.249 66.976 45.006 -> 66.976 -> 3817.623 MByte/s p20 acyclic-2dim-all : 41.347 46.346 35.045 -> 46.346 -> 2641.727 MByte/s p21 acyclic-3dim-all : 41.507 49.119 37.312 -> 49.119 -> 2799.761 MByte/s p22 cyclic-1dim-all : 54.929 77.004 45.766 -> 77.004 -> 4389.213 MByte/s p23 cyclic-2dim-all : 42.809 46.560 36.103 -> 46.560 -> 2653.927 MByte/s p24 cyclic-3dim-all : 45.348 52.630 40.526 -> 52.630 -> 2999.918 MByte/s log_avg of all rings : 56.734 72.551 45.491 || 72.551 -> 4135.385 MByte/s log_avg of all random : 35.368 37.559 31.205 || 37.711 -> 2149.528 MByte/s log_avg(ring,random) : 44.795 52.201 37.677 ||( 52.306 -> 2981.464)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-28*2&+1 : 72.329 72.566 72.248 -> 72.566 -> 4136.249 MByte/s p01 ring-14*4&+1 : 75.443 75.901 75.123 -> 75.901 -> 4326.338 MByte/s p02 ring-7*8&+1 : 67.653 67.403 67.519 -> 67.653 -> 3856.198 MByte/s p03 ring-3*19fix : 64.057 65.145 64.571 -> 65.145 -> 3713.257 MByte/s p04 ring-2*28&+1 : 77.200 76.526 76.024 -> 77.200 -> 4400.415 MByte/s p05 ring-1*57fix : 74.812 76.451 76.215 -> 76.451 -> 4357.724 MByte/s p06 random-cyc-1dim : 41.931 41.273 41.227 -> 41.931 -> 2390.059 MByte/s p07 random-cyc-1dim : 44.986 45.198 45.604 -> 45.604 -> 2599.437 MByte/s p08 random-cyc-1dim : 38.051 38.351 38.361 -> 38.361 -> 2186.588 MByte/s p09 random-cyc-1dim : 41.750 41.314 41.536 -> 41.750 -> 2379.773 MByte/s p10 random-cyc-1dim : 38.985 39.172 39.165 -> 39.172 -> 2232.782 MByte/s p11 random-cyc-1dim : 41.189 40.938 41.222 -> 41.222 -> 2349.647 MByte/s p12 random-cyc-1dim : 36.840 36.321 36.327 -> 36.840 -> 2099.881 MByte/s p13 random-cyc-1dim : 35.595 36.258 36.046 -> 36.258 -> 2066.680 MByte/s p14 random-cyc-1dim : 37.434 37.073 36.966 -> 37.434 -> 2133.721 MByte/s p15 random-cyc-1dim : 34.216 34.278 34.240 -> 34.278 -> 1953.869 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 26.693 26.580 26.361 -> 26.693 -> 1521.511 MByte/s p17 best bi-section : 69.442 69.605 69.795 -> 69.795 -> 3978.340 MByte/s p18 worst bi-section : 20.599 20.404 20.469 -> 20.599 -> 1174.160 MByte/s p19 acyclic-1dim-all : 66.077 65.191 66.325 -> 66.325 -> 3780.504 MByte/s p20 acyclic-2dim-all : 48.896 49.963 49.293 -> 49.963 -> 2847.891 MByte/s p21 acyclic-3dim-all : 51.869 51.695 52.026 -> 52.026 -> 2965.479 MByte/s p22 cyclic-1dim-all : 76.010 76.254 76.531 -> 76.531 -> 4362.277 MByte/s p23 cyclic-2dim-all : 49.919 49.987 50.381 -> 50.381 -> 2871.717 MByte/s p24 cyclic-3dim-all : 54.293 54.164 54.373 -> 54.373 -> 3099.281 MByte/s log_avg of all rings : 71.762 72.186 71.808 || 72.337 -> 4123.182 MByte/s log_avg of all random : 38.972 38.899 38.940 || 39.158 -> 2231.999 MByte/s log_avg(ring,random) : 52.884 52.990 52.879 ||( 53.222 -> 3033.635)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-28*2&+1 p00 method 0 : 0.053 0.824 9.807 49.987 179.123 207.509 -> 67.587 -> 3852.451 MByte/s p00 method 1 : 0.018 0.312 4.734 57.446 180.618 208.335 -> 71.170 -> 4056.693 MByte/s p00 method 2 : 0.024 0.386 4.748 26.428 127.529 121.516 -> 49.119 -> 2799.781 MByte/s p01 ring-14*4&+1 p01 method 0 : 0.051 0.834 9.728 50.090 165.544 153.691 -> 58.405 -> 3329.103 MByte/s p01 method 1 : 0.030 0.521 7.744 80.222 185.448 195.150 -> 75.838 -> 4322.783 MByte/s p01 method 2 : 0.021 0.341 4.477 29.714 115.595 112.796 -> 44.588 -> 2541.492 MByte/s p02 ring-7*8&+1 p02 method 0 : 0.052 0.821 9.456 48.490 135.348 137.615 -> 53.611 -> 3055.823 MByte/s p02 method 1 : 0.030 0.524 7.783 78.306 152.687 188.966 -> 68.427 -> 3900.348 MByte/s p02 method 2 : 0.022 0.342 4.422 30.080 117.298 117.196 -> 44.803 -> 2553.794 MByte/s p03 ring-3*19fix p03 method 0 : 0.047 0.774 8.436 42.221 136.452 130.268 -> 52.789 -> 3008.978 MByte/s p03 method 1 : 0.030 0.513 7.636 76.619 148.047 186.143 -> 65.671 -> 3743.237 MByte/s p03 method 2 : 0.021 0.341 4.455 29.880 114.325 103.726 -> 43.533 -> 2481.357 MByte/s p04 ring-2*28&+1 p04 method 0 : 0.048 0.768 8.445 39.876 143.806 130.262 -> 55.035 -> 3137.018 MByte/s p04 method 1 : 0.030 0.522 7.771 79.971 193.009 212.451 -> 77.818 -> 4435.605 MByte/s p04 method 2 : 0.022 0.343 4.508 29.485 125.204 103.184 -> 45.648 -> 2601.927 MByte/s p05 ring-1*57fix p05 method 0 : 0.048 0.767 8.365 39.022 141.768 131.783 -> 54.241 -> 3091.738 MByte/s p05 method 1 : 0.030 0.519 7.712 78.716 191.523 210.973 -> 77.265 -> 4404.093 MByte/s p05 method 2 : 0.022 0.345 4.457 29.896 123.862 102.702 -> 45.451 -> 2590.721 MByte/s p06 random-cyc-1dim p06 method 0 : 0.044 0.706 7.799 38.198 89.234 80.339 -> 37.360 -> 2129.503 MByte/s p06 method 1 : 0.029 0.510 7.655 70.067 91.271 77.468 -> 40.548 -> 2311.255 MByte/s p06 method 2 : 0.021 0.337 4.341 28.151 83.255 83.406 -> 33.213 -> 1893.163 MByte/s p07 random-cyc-1dim p07 method 0 : 0.045 0.717 7.790 38.090 99.597 87.473 -> 40.058 -> 2283.330 MByte/s p07 method 1 : 0.030 0.515 7.717 72.449 100.106 89.547 -> 45.149 -> 2573.496 MByte/s p07 method 2 : 0.021 0.336 4.336 29.406 90.956 84.608 -> 35.377 -> 2016.486 MByte/s p08 random-cyc-1dim p08 method 0 : 0.044 0.717 7.822 36.996 83.051 78.727 -> 35.121 -> 2001.898 MByte/s p08 method 1 : 0.030 0.510 7.666 68.106 80.935 65.501 -> 37.206 -> 2120.762 MByte/s p08 method 2 : 0.021 0.334 4.373 29.308 76.957 66.406 -> 30.905 -> 1761.604 MByte/s p09 random-cyc-1dim p09 method 0 : 0.045 0.701 7.688 37.499 95.008 74.923 -> 37.568 -> 2141.392 MByte/s p09 method 1 : 0.029 0.503 7.561 68.793 81.478 50.139 -> 37.542 -> 2139.875 MByte/s p09 method 2 : 0.021 0.335 4.357 29.023 78.107 70.422 -> 31.947 -> 1820.970 MByte/s p10 random-cyc-1dim p10 method 0 : 0.044 0.704 7.722 37.504 89.676 65.284 -> 35.266 -> 2010.134 MByte/s p10 method 1 : 0.030 0.517 7.655 69.556 86.990 45.569 -> 37.238 -> 2122.593 MByte/s p10 method 2 : 0.021 0.323 4.339 29.314 79.588 54.415 -> 30.039 -> 1712.242 MByte/s p11 random-cyc-1dim p11 method 0 : 0.044 0.700 7.723 37.633 93.280 81.922 -> 36.951 -> 2106.179 MByte/s p11 method 1 : 0.029 0.509 7.609 70.098 92.860 78.090 -> 41.268 -> 2352.257 MByte/s p11 method 2 : 0.021 0.334 4.305 29.194 84.025 76.502 -> 33.756 -> 1924.119 MByte/s p12 random-cyc-1dim p12 method 0 : 0.045 0.729 7.843 36.781 78.428 68.775 -> 33.132 -> 1888.531 MByte/s p12 method 1 : 0.030 0.512 7.686 65.911 79.996 71.140 -> 36.958 -> 2106.617 MByte/s p12 method 2 : 0.021 0.337 4.326 28.350 71.637 70.510 -> 30.106 -> 1716.052 MByte/s p13 random-cyc-1dim p13 method 0 : 0.045 0.725 7.803 37.116 78.769 79.444 -> 33.760 -> 1924.297 MByte/s p13 method 1 : 0.030 0.505 7.581 64.127 69.328 50.168 -> 32.444 -> 1849.325 MByte/s p13 method 2 : 0.021 0.337 4.259 28.712 69.298 71.322 -> 28.457 -> 1622.040 MByte/s p14 random-cyc-1dim p14 method 0 : 0.044 0.705 7.789 36.859 81.529 82.671 -> 34.652 -> 1975.189 MByte/s p14 method 1 : 0.030 0.519 7.678 64.670 75.781 52.390 -> 35.028 -> 1996.615 MByte/s p14 method 2 : 0.021 0.336 4.332 29.080 72.836 67.419 -> 30.277 -> 1725.794 MByte/s p15 random-cyc-1dim p15 method 0 : 0.042 0.670 7.735 36.523 71.875 61.553 -> 30.706 -> 1750.242 MByte/s p15 method 1 : 0.030 0.510 7.678 63.658 71.368 58.287 -> 33.845 -> 1929.160 MByte/s p15 method 2 : 0.020 0.320 4.286 28.966 69.816 62.755 -> 28.688 -> 1635.208 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.043 0.628 7.595 34.971 53.171 44.819 -> 24.248 -> 1382.116 MByte/s p16 method 1 : 0.030 0.509 7.748 50.802 49.356 33.062 -> 24.794 -> 1413.277 MByte/s p16 method 2 : 0.020 0.313 4.239 28.508 50.546 49.471 -> 22.382 -> 1275.770 MByte/s p17 best bi-section p17 method 0 : 0.032 0.485 5.503 27.419 128.135 159.956 -> 50.524 -> 2879.858 MByte/s p17 method 1 : 0.018 0.304 4.636 55.583 177.573 204.056 -> 69.900 -> 3984.276 MByte/s p17 method 2 : 0.014 0.231 3.378 26.729 123.129 145.152 -> 47.717 -> 2719.888 MByte/s p18 worst bi-section p18 method 0 : 0.023 0.368 4.393 24.652 37.750 39.818 -> 17.800 -> 1014.613 MByte/s p18 method 1 : 0.018 0.296 4.524 32.871 39.039 54.921 -> 20.685 -> 1179.040 MByte/s p18 method 2 : 0.014 0.227 3.265 23.814 36.650 35.335 -> 17.238 -> 982.538 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.047 0.757 8.303 40.308 140.307 138.628 -> 54.249 -> 3092.209 MByte/s p19 method 1 : 0.030 0.519 7.699 79.581 165.712 140.365 -> 66.976 -> 3817.623 MByte/s p19 method 2 : 0.021 0.335 4.335 28.845 123.685 95.360 -> 45.006 -> 2565.344 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.035 0.554 6.196 29.535 105.465 103.822 -> 41.347 -> 2356.787 MByte/s p20 method 1 : 0.038 0.667 9.536 80.164 103.497 83.346 -> 46.346 -> 2641.727 MByte/s p20 method 2 : 0.017 0.276 3.545 24.580 92.745 78.990 -> 35.045 -> 1997.586 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.032 0.496 5.531 25.916 106.221 118.892 -> 41.507 -> 2365.884 MByte/s p21 method 1 : 0.045 0.791 11.135 79.975 108.147 87.976 -> 49.119 -> 2799.761 MByte/s p21 method 2 : 0.016 0.262 3.399 23.529 99.090 91.530 -> 37.312 -> 2126.806 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.047 0.764 8.212 38.538 141.021 143.422 -> 54.929 -> 3130.979 MByte/s p22 method 1 : 0.030 0.514 7.605 78.042 192.002 210.728 -> 77.004 -> 4389.213 MByte/s p22 method 2 : 0.022 0.348 4.522 29.131 124.220 101.526 -> 45.766 -> 2608.652 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.040 0.631 6.840 32.721 107.759 107.501 -> 42.809 -> 2440.111 MByte/s p23 method 1 : 0.044 0.771 11.114 77.785 101.458 85.708 -> 46.560 -> 2653.927 MByte/s p23 method 2 : 0.021 0.340 4.350 29.741 93.489 81.703 -> 36.103 -> 2057.873 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.037 0.606 6.489 31.048 115.813 114.145 -> 45.348 -> 2584.810 MByte/s p24 method 1 : 0.049 0.865 12.114 82.520 113.143 106.785 -> 52.630 -> 2999.918 MByte/s p24 method 2 : 0.021 0.338 4.226 29.274 106.920 94.817 -> 40.526 -> 2309.999 MByte/s log_avg of all rings - ring, method 0 : 0.050 0.798 9.017 44.699 149.500 146.322 || 56.734 -> 3233.865 MByte/s - ring, method 1 : 0.028 0.477 7.123 74.722 174.247 200.051 || 72.551 -> 4135.385 MByte/s - ring, method 2 : 0.022 0.349 4.510 29.218 120.528 109.940 || 45.491 -> 2592.988 MByte/s log_avg of all random - random, method 0 : 0.044 0.707 7.771 37.316 85.646 75.681 || 35.368 -> 2015.955 MByte/s - random, method 1 : 0.030 0.511 7.649 67.684 82.496 62.319 || 37.559 -> 2140.851 MByte/s - random, method 2 : 0.021 0.333 4.325 28.948 77.368 70.240 || 31.205 -> 1778.667 MByte/s log_avg(ring,random) - average, method 0 : 0.047 0.751 8.371 40.841 113.155 105.232 || 44.795 -> 2553.297 MByte/s - average, method 1 : 0.029 0.494 7.381 71.116 119.894 111.656 || 52.201 -> 2975.440 MByte/s - average, method 2 : 0.021 0.341 4.417 29.082 96.566 87.876 || 37.677 -> 2147.571 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 2.674 0.047 0.050 0.044 0.047 0.029 0.021 2 5.372 0.094 0.100 0.089 0.094 0.057 0.043 4 10.768 0.189 0.200 0.179 0.189 0.126 0.086 8 21.526 0.378 0.403 0.354 0.378 0.250 0.172 16 42.809 0.751 0.798 0.707 0.751 0.494 0.341 32 65.541 1.150 1.249 1.058 1.150 0.972 0.592 64 119.885 2.103 2.275 1.944 2.092 1.870 1.125 128 259.250 4.548 4.921 4.203 4.548 3.865 2.334 256 477.160 8.371 9.017 7.771 8.371 7.381 4.417 512 871.265 15.285 15.998 14.605 14.919 14.141 8.094 1024 1564.560 27.448 27.725 27.175 22.927 26.500 13.904 2048 2651.369 46.515 46.755 46.277 31.505 45.996 21.401 4096 4053.609 71.116 74.722 67.684 40.841 71.116 29.082 8192 5226.506 91.693 106.913 78.640 63.865 91.693 60.496 16384 6096.921 106.964 138.950 82.340 86.636 106.742 80.778 32768 6687.481 117.324 162.781 84.561 102.604 116.713 91.469 65536 6979.771 122.452 174.247 86.053 113.155 119.894 96.566 131072 7116.425 124.850 180.664 86.279 114.768 119.572 98.754 262144 7198.621 126.292 188.383 84.666 113.561 120.352 97.905 524288 7086.541 124.325 193.648 79.819 108.318 116.701 92.712 1048576 7053.655 123.748 200.051 76.549 105.232 111.656 87.876 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-28*2&+1 : 0.053 0.824 9.807 57.446 180.618 208.335 -> 72.762 -> 4147.443 MByte/s p01 ring-14*4&+1 : 0.051 0.834 9.728 80.222 185.448 195.150 -> 76.169 -> 4341.660 MByte/s p02 ring-7*8&+1 : 0.052 0.821 9.456 78.306 152.687 188.966 -> 68.713 -> 3916.659 MByte/s p03 ring-3*19fix : 0.047 0.774 8.436 76.619 148.047 186.143 -> 65.808 -> 3751.078 MByte/s p04 ring-2*28&+1 : 0.048 0.768 8.445 79.971 193.009 212.451 -> 77.914 -> 4441.120 MByte/s p05 ring-1*57fix : 0.048 0.767 8.365 78.716 191.523 210.973 -> 77.367 -> 4409.905 MByte/s p06 random-cyc-1dim : 0.044 0.706 7.799 70.067 91.271 83.406 -> 42.213 -> 2406.122 MByte/s p07 random-cyc-1dim : 0.045 0.717 7.790 72.449 100.106 89.547 -> 45.754 -> 2608.003 MByte/s p08 random-cyc-1dim : 0.044 0.717 7.822 68.106 83.051 78.727 -> 39.076 -> 2227.349 MByte/s p09 random-cyc-1dim : 0.045 0.701 7.688 68.793 95.008 74.923 -> 42.170 -> 2403.666 MByte/s p10 random-cyc-1dim : 0.044 0.704 7.722 69.556 89.676 65.284 -> 39.568 -> 2255.377 MByte/s p11 random-cyc-1dim : 0.044 0.700 7.723 70.098 93.280 81.922 -> 41.752 -> 2379.868 MByte/s p12 random-cyc-1dim : 0.045 0.729 7.843 65.911 79.996 71.140 -> 37.154 -> 2117.805 MByte/s p13 random-cyc-1dim : 0.045 0.725 7.803 64.127 78.769 79.444 -> 36.791 -> 2097.097 MByte/s p14 random-cyc-1dim : 0.044 0.705 7.789 64.670 81.529 82.671 -> 38.204 -> 2177.640 MByte/s p15 random-cyc-1dim : 0.042 0.670 7.735 63.658 71.875 62.755 -> 34.691 -> 1977.414 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.043 0.628 7.748 50.802 53.171 49.471 -> 26.850 -> 1530.467 MByte/s p17 best bi-section : 0.032 0.485 5.503 55.583 177.573 204.056 -> 70.021 -> 3991.206 MByte/s p18 worst bi-section : 0.023 0.368 4.524 32.871 39.039 54.921 -> 20.697 -> 1179.752 MByte/s p19 acyclic-1dim-all : 0.047 0.757 8.303 79.581 165.712 140.365 -> 67.075 -> 3823.272 MByte/s p20 acyclic-2dim-all : 0.038 0.667 9.536 80.164 105.465 103.822 -> 50.072 -> 2854.081 MByte/s p21 acyclic-3dim-all : 0.045 0.791 11.135 79.975 108.147 118.892 -> 52.261 -> 2978.874 MByte/s p22 cyclic-1dim-all : 0.047 0.764 8.212 78.042 192.002 210.728 -> 77.118 -> 4395.701 MByte/s p23 cyclic-2dim-all : 0.044 0.771 11.114 77.785 107.759 107.501 -> 50.573 -> 2882.649 MByte/s p24 cyclic-3dim-all : 0.049 0.865 12.114 82.520 115.813 114.145 -> 54.818 -> 3124.599 MByte/s log_avg of all rings : 0.050 0.798 9.017 74.722 174.247 200.051 || 72.979 -> 4159.792 MByte/s log_avg of all random : 0.044 0.707 7.771 67.684 86.053 76.549 || 39.618 -> 2258.213 MByte/s log_avg(ring,random) : 0.047 0.751 8.371 71.116 122.452 123.748 || 53.770 -> 3064.914 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 3064.914 MByte/s on 57 processes ( = 53.770 MByte/s * 57 processes) system parameters : 57 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 = 3064.914 MB/s = 53.770 * 57 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E