b_eff = 1893.872 MB/s = 59.183 * 32 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 = 32 1-dim-paterns: size = 32 2-dim-paterns: size = 8 * 4 3-dim-paterns: size = 4 * 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 = 138.540 sec sum of max elapsed time per entries above = 141.228 sec difference = -2.689 sec = 1.9% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-16*2fix => 1 sendrecv_calls with 32 messages, i.e. msgs/used node, all nodes are used p01 ring-8*4fix => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p02 ring-4*8fix => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p03 ring-2*16fix => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p04 ring-1*32fix => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p05 ring-1*32fix => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 32 messages, i.e. msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 32 messages, i.e. msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 62 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 104 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 128 messages, i.e. msgs/used node, all nodes are used p22 cyclic-1dim-all => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 128 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 5 sendrecv_calls with 160 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-16*2fix : 72.242 78.035 64.837 -> 78.035 -> 2497.129 MByte/s p01 ring-8*4fix : 61.305 80.983 46.262 -> 80.983 -> 2591.457 MByte/s p02 ring-4*8fix : 53.775 73.260 45.254 -> 73.260 -> 2344.330 MByte/s p03 ring-2*16fix : 52.215 69.307 44.647 -> 69.307 -> 2217.811 MByte/s p04 ring-1*32fix : 56.334 84.153 46.526 -> 84.153 -> 2692.904 MByte/s p05 ring-1*32fix : 55.614 83.985 47.611 -> 83.985 -> 2687.517 MByte/s p06 random-cyc-1dim : 44.384 50.227 37.661 -> 50.227 -> 1607.270 MByte/s p07 random-cyc-1dim : 33.261 37.191 30.253 -> 37.191 -> 1190.119 MByte/s p08 random-cyc-1dim : 33.380 35.081 28.814 -> 35.081 -> 1122.607 MByte/s p09 random-cyc-1dim : 34.544 40.018 31.245 -> 40.018 -> 1280.589 MByte/s p10 random-cyc-1dim : 42.207 50.749 37.617 -> 50.749 -> 1623.976 MByte/s p11 random-cyc-1dim : 42.496 43.306 34.879 -> 43.306 -> 1385.786 MByte/s p12 random-cyc-1dim : 41.312 49.671 36.167 -> 49.671 -> 1589.465 MByte/s p13 random-cyc-1dim : 39.092 41.743 33.398 -> 41.743 -> 1335.771 MByte/s p14 random-cyc-1dim : 30.195 33.286 26.219 -> 33.286 -> 1065.165 MByte/s p15 random-cyc-1dim : 42.320 48.220 36.752 -> 48.220 -> 1543.055 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 31.224 34.959 27.531 -> 34.959 -> 1118.699 MByte/s p17 best bi-section : 51.510 78.041 49.148 -> 78.041 -> 2497.318 MByte/s p18 worst bi-section : 26.536 33.208 25.920 -> 33.208 -> 1062.642 MByte/s p19 acyclic-1dim-all : 53.898 70.544 46.070 -> 70.544 -> 2257.404 MByte/s p20 acyclic-2dim-all : 42.365 49.054 35.203 -> 49.054 -> 1569.742 MByte/s p21 acyclic-3dim-all : 43.128 50.275 37.776 -> 50.275 -> 1608.790 MByte/s p22 cyclic-1dim-all : 55.859 83.369 46.442 -> 83.369 -> 2667.798 MByte/s p23 cyclic-2dim-all : 46.506 50.862 38.442 -> 50.862 -> 1627.590 MByte/s p24 cyclic-3dim-all : 46.435 56.013 42.204 -> 56.013 -> 1792.406 MByte/s log_avg of all rings : 58.224 78.091 48.751 || 78.091 -> 2498.907 MByte/s log_avg of all random : 38.014 42.490 33.074 || 42.490 -> 1359.682 MByte/s log_avg(ring,random) : 47.046 57.603 40.155 ||( 57.603 -> 1843.290)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-16*2fix : 78.346 77.383 78.371 -> 78.371 -> 2507.861 MByte/s p01 ring-8*4fix : 80.748 80.586 78.145 -> 80.748 -> 2583.950 MByte/s p02 ring-4*8fix : 71.860 71.251 71.208 -> 71.860 -> 2299.505 MByte/s p03 ring-2*16fix : 68.372 68.302 68.530 -> 68.530 -> 2192.954 MByte/s p04 ring-1*32fix : 81.404 82.540 82.971 -> 82.971 -> 2655.072 MByte/s p05 ring-1*32fix : 83.135 82.336 82.365 -> 83.135 -> 2660.326 MByte/s p06 random-cyc-1dim : 51.367 52.328 51.718 -> 52.328 -> 1674.507 MByte/s p07 random-cyc-1dim : 37.678 38.061 38.392 -> 38.392 -> 1228.553 MByte/s p08 random-cyc-1dim : 38.230 37.940 38.162 -> 38.230 -> 1223.350 MByte/s p09 random-cyc-1dim : 40.126 40.408 40.711 -> 40.711 -> 1302.740 MByte/s p10 random-cyc-1dim : 50.802 51.153 50.576 -> 51.153 -> 1636.897 MByte/s p11 random-cyc-1dim : 46.326 47.324 47.648 -> 47.648 -> 1524.740 MByte/s p12 random-cyc-1dim : 49.351 49.302 49.288 -> 49.351 -> 1579.217 MByte/s p13 random-cyc-1dim : 44.061 44.462 45.032 -> 45.032 -> 1441.009 MByte/s p14 random-cyc-1dim : 33.884 34.480 34.090 -> 34.480 -> 1103.346 MByte/s p15 random-cyc-1dim : 49.657 50.300 49.579 -> 50.300 -> 1609.600 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 35.889 35.947 35.842 -> 35.947 -> 1150.305 MByte/s p17 best bi-section : 77.403 77.759 77.860 -> 77.860 -> 2491.525 MByte/s p18 worst bi-section : 33.346 33.228 33.745 -> 33.745 -> 1079.844 MByte/s p19 acyclic-1dim-all : 68.750 68.968 69.208 -> 69.208 -> 2214.671 MByte/s p20 acyclic-2dim-all : 52.502 52.403 52.520 -> 52.520 -> 1680.634 MByte/s p21 acyclic-3dim-all : 53.044 51.820 53.731 -> 53.731 -> 1719.397 MByte/s p22 cyclic-1dim-all : 82.514 82.946 81.156 -> 82.946 -> 2654.260 MByte/s p23 cyclic-2dim-all : 55.648 55.435 55.259 -> 55.648 -> 1780.744 MByte/s p24 cyclic-3dim-all : 57.310 57.579 57.686 -> 57.686 -> 1845.941 MByte/s log_avg of all rings : 77.118 76.865 76.740 || 77.398 -> 2476.721 MByte/s log_avg of all random : 43.728 44.144 44.110 || 44.340 -> 1418.876 MByte/s log_avg(ring,random) : 58.071 58.251 58.181 ||( 58.582 -> 1874.609)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-16*2fix p00 method 0 : 0.055 0.868 10.365 51.941 179.552 207.911 -> 72.242 -> 2311.743 MByte/s p00 method 1 : 0.029 0.511 7.704 79.530 191.311 208.973 -> 78.035 -> 2497.129 MByte/s p00 method 2 : 0.025 0.400 5.284 26.649 169.446 206.554 -> 64.837 -> 2074.797 MByte/s p01 ring-8*4fix p01 method 0 : 0.054 0.843 9.950 50.191 172.571 153.779 -> 61.305 -> 1961.757 MByte/s p01 method 1 : 0.044 0.791 11.446 98.235 190.714 197.423 -> 80.983 -> 2591.457 MByte/s p01 method 2 : 0.022 0.348 4.664 29.669 117.083 113.287 -> 46.262 -> 1480.376 MByte/s p02 ring-4*8fix p02 method 0 : 0.053 0.824 9.479 48.703 135.132 133.867 -> 53.775 -> 1720.804 MByte/s p02 method 1 : 0.044 0.788 11.467 98.757 153.422 190.005 -> 73.260 -> 2344.330 MByte/s p02 method 2 : 0.022 0.342 4.468 30.050 118.224 113.798 -> 45.254 -> 1448.126 MByte/s p03 ring-2*16fix p03 method 0 : 0.049 0.786 8.564 41.435 130.715 132.491 -> 52.215 -> 1670.869 MByte/s p03 method 1 : 0.044 0.781 11.312 91.644 146.816 171.532 -> 69.307 -> 2217.811 MByte/s p03 method 2 : 0.022 0.343 4.485 30.091 116.280 111.512 -> 44.647 -> 1428.699 MByte/s p04 ring-1*32fix p04 method 0 : 0.048 0.779 8.624 40.381 153.664 144.377 -> 56.334 -> 1802.687 MByte/s p04 method 1 : 0.043 0.776 11.317 97.242 198.840 212.248 -> 84.153 -> 2692.904 MByte/s p04 method 2 : 0.022 0.344 4.472 29.995 125.606 103.340 -> 46.526 -> 1488.819 MByte/s p05 ring-1*32fix p05 method 0 : 0.048 0.764 8.724 41.092 142.935 133.800 -> 55.614 -> 1779.639 MByte/s p05 method 1 : 0.043 0.775 11.342 97.745 198.683 211.079 -> 83.985 -> 2687.517 MByte/s p05 method 2 : 0.022 0.346 4.511 30.225 126.818 118.962 -> 47.611 -> 1523.546 MByte/s p06 random-cyc-1dim p06 method 0 : 0.046 0.752 8.036 38.358 108.170 112.009 -> 44.384 -> 1420.301 MByte/s p06 method 1 : 0.043 0.768 11.186 85.375 106.856 98.847 -> 50.227 -> 1607.270 MByte/s p06 method 2 : 0.022 0.340 4.420 29.471 95.027 89.751 -> 37.661 -> 1205.141 MByte/s p07 random-cyc-1dim p07 method 0 : 0.045 0.708 7.864 36.319 79.669 63.354 -> 33.261 -> 1064.344 MByte/s p07 method 1 : 0.044 0.785 11.411 69.965 77.244 60.503 -> 37.191 -> 1190.119 MByte/s p07 method 2 : 0.021 0.335 4.352 29.458 74.227 66.627 -> 30.253 -> 968.104 MByte/s p08 random-cyc-1dim p08 method 0 : 0.045 0.715 7.800 36.788 79.457 75.478 -> 33.380 -> 1068.156 MByte/s p08 method 1 : 0.044 0.773 11.300 69.499 73.031 44.363 -> 35.081 -> 1122.607 MByte/s p08 method 2 : 0.021 0.336 4.346 29.325 72.015 67.364 -> 28.814 -> 922.050 MByte/s p09 random-cyc-1dim p09 method 0 : 0.046 0.736 7.917 37.187 87.742 67.857 -> 34.544 -> 1105.405 MByte/s p09 method 1 : 0.045 0.778 11.402 73.372 83.494 68.263 -> 40.018 -> 1280.589 MByte/s p09 method 2 : 0.021 0.339 4.366 29.215 76.363 74.095 -> 31.245 -> 999.841 MByte/s p10 random-cyc-1dim p10 method 0 : 0.047 0.762 7.941 38.447 104.496 98.941 -> 42.207 -> 1350.632 MByte/s p10 method 1 : 0.044 0.767 11.184 85.802 110.987 82.624 -> 50.749 -> 1623.976 MByte/s p10 method 2 : 0.021 0.338 4.393 29.652 96.828 90.201 -> 37.617 -> 1203.738 MByte/s p11 random-cyc-1dim p11 method 0 : 0.046 0.730 7.942 38.175 103.552 103.167 -> 42.496 -> 1359.862 MByte/s p11 method 1 : 0.044 0.774 11.267 73.373 92.361 74.772 -> 43.306 -> 1385.786 MByte/s p11 method 2 : 0.021 0.338 4.395 29.481 90.903 69.786 -> 34.879 -> 1116.118 MByte/s p12 random-cyc-1dim p12 method 0 : 0.046 0.755 8.187 38.760 101.694 98.253 -> 41.312 -> 1321.986 MByte/s p12 method 1 : 0.044 0.774 11.247 82.985 103.340 98.513 -> 49.671 -> 1589.465 MByte/s p12 method 2 : 0.021 0.340 4.417 29.555 95.050 78.379 -> 36.167 -> 1157.346 MByte/s p13 random-cyc-1dim p13 method 0 : 0.045 0.733 7.885 37.906 97.464 87.636 -> 39.092 -> 1250.940 MByte/s p13 method 1 : 0.043 0.769 11.144 80.551 91.151 51.412 -> 41.743 -> 1335.771 MByte/s p13 method 2 : 0.021 0.334 4.380 29.302 84.857 80.761 -> 33.398 -> 1068.733 MByte/s p14 random-cyc-1dim p14 method 0 : 0.044 0.712 7.917 36.330 70.881 65.146 -> 30.195 -> 966.247 MByte/s p14 method 1 : 0.044 0.761 11.085 63.077 67.763 47.003 -> 33.286 -> 1065.165 MByte/s p14 method 2 : 0.021 0.335 4.326 29.020 62.995 49.351 -> 26.219 -> 839.002 MByte/s p15 random-cyc-1dim p15 method 0 : 0.046 0.749 8.101 38.523 110.040 86.970 -> 42.320 -> 1354.228 MByte/s p15 method 1 : 0.044 0.764 11.172 84.112 101.606 86.045 -> 48.220 -> 1543.055 MByte/s p15 method 2 : 0.021 0.340 4.423 29.636 98.028 79.015 -> 36.752 -> 1176.061 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.044 0.750 7.980 36.234 72.209 60.792 -> 31.224 -> 999.164 MByte/s p16 method 1 : 0.044 0.763 11.157 65.194 69.733 49.579 -> 34.959 -> 1118.699 MByte/s p16 method 2 : 0.021 0.335 4.304 28.952 65.842 56.892 -> 27.531 -> 880.978 MByte/s p17 best bi-section p17 method 0 : 0.030 0.491 5.496 27.978 131.475 162.605 -> 51.510 -> 1648.328 MByte/s p17 method 1 : 0.030 0.505 7.567 78.645 190.668 209.202 -> 78.041 -> 2497.318 MByte/s p17 method 2 : 0.015 0.237 3.465 27.722 126.386 161.904 -> 49.148 -> 1572.734 MByte/s p18 worst bi-section p18 method 0 : 0.024 0.382 4.631 26.657 66.589 55.881 -> 26.536 -> 849.156 MByte/s p18 method 1 : 0.028 0.491 7.362 55.233 62.751 87.341 -> 33.208 -> 1062.642 MByte/s p18 method 2 : 0.014 0.232 3.429 26.785 64.502 56.344 -> 25.920 -> 829.446 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.047 0.746 8.634 39.630 139.558 127.740 -> 53.898 -> 1724.721 MByte/s p19 method 1 : 0.044 0.774 11.227 97.616 165.720 141.603 -> 70.544 -> 2257.404 MByte/s p19 method 2 : 0.021 0.330 4.270 29.688 125.696 118.427 -> 46.070 -> 1474.246 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.034 0.540 6.005 27.571 109.727 115.259 -> 42.365 -> 1355.680 MByte/s p20 method 1 : 0.051 0.915 12.690 83.178 106.185 80.592 -> 49.054 -> 1569.742 MByte/s p20 method 2 : 0.018 0.281 3.628 24.912 93.617 81.443 -> 35.203 -> 1126.512 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.031 0.491 5.526 25.242 109.811 131.384 -> 43.128 -> 1380.097 MByte/s p21 method 1 : 0.049 0.873 12.131 81.184 107.053 101.083 -> 50.275 -> 1608.790 MByte/s p21 method 2 : 0.016 0.254 3.301 23.321 101.289 94.863 -> 37.776 -> 1208.823 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.049 0.792 8.794 39.069 143.590 144.267 -> 55.859 -> 1787.486 MByte/s p22 method 1 : 0.042 0.746 10.891 95.652 197.437 212.852 -> 83.369 -> 2667.798 MByte/s p22 method 2 : 0.022 0.349 4.525 29.948 129.785 101.758 -> 46.442 -> 1486.150 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.041 0.645 6.934 32.646 118.184 120.172 -> 46.506 -> 1488.178 MByte/s p23 method 1 : 0.058 1.049 14.604 86.745 111.860 83.079 -> 50.862 -> 1627.590 MByte/s p23 method 2 : 0.022 0.341 4.428 29.976 99.441 84.806 -> 38.442 -> 1230.146 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.041 0.630 7.346 31.830 117.789 120.262 -> 46.435 -> 1485.917 MByte/s p24 method 1 : 0.063 1.138 15.665 89.749 117.448 113.063 -> 56.013 -> 1792.406 MByte/s p24 method 2 : 0.022 0.347 4.466 29.679 109.132 103.501 -> 42.204 -> 1350.520 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.810 9.259 45.377 151.353 149.027 || 58.224 -> 1863.167 MByte/s - ring, method 1 : 0.041 0.729 10.661 93.592 178.611 197.992 || 78.091 -> 2498.907 MByte/s - ring, method 2 : 0.022 0.353 4.639 29.418 127.738 124.086 || 48.751 -> 1560.027 MByte/s log_avg of all random - random, method 0 : 0.046 0.735 7.958 37.669 93.351 84.288 || 38.014 -> 1216.436 MByte/s - random, method 1 : 0.044 0.771 11.239 76.428 89.630 68.552 || 42.490 -> 1359.682 MByte/s - random, method 2 : 0.021 0.337 4.382 29.411 83.764 73.558 || 33.074 -> 1058.379 MByte/s log_avg(ring,random) - average, method 0 : 0.048 0.772 8.584 41.343 118.866 112.077 || 47.046 -> 1505.465 MByte/s - average, method 1 : 0.042 0.750 10.946 84.576 126.526 116.502 || 57.603 -> 1843.290 MByte/s - average, method 2 : 0.022 0.345 4.508 29.415 103.440 95.538 || 40.155 -> 1284.951 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 1.545 0.048 0.051 0.046 0.048 0.042 0.022 2 3.087 0.096 0.103 0.091 0.096 0.084 0.043 4 6.520 0.204 0.208 0.199 0.194 0.193 0.087 8 12.964 0.405 0.415 0.396 0.387 0.384 0.174 16 25.320 0.791 0.812 0.771 0.772 0.750 0.345 32 48.763 1.524 1.526 1.522 1.186 1.479 0.604 64 90.782 2.837 2.840 2.834 2.151 2.765 1.149 128 191.916 5.997 6.002 5.993 4.641 5.835 2.384 256 359.050 11.220 11.201 11.239 8.584 10.946 4.508 512 664.048 20.752 20.612 20.892 15.179 20.446 8.217 1024 1179.655 36.864 36.608 37.123 23.711 36.795 14.205 2048 1883.392 58.856 60.792 56.982 32.130 58.856 21.839 4096 2706.424 84.576 93.592 76.428 41.343 84.576 29.415 8192 3348.598 104.644 127.605 85.814 65.763 104.644 62.208 16384 3772.826 117.901 154.764 89.818 88.573 117.901 82.878 32768 4021.183 125.662 172.227 91.687 107.669 125.285 96.845 65536 4147.838 129.620 178.611 94.067 118.866 126.526 103.440 131072 4217.248 131.789 184.813 93.978 120.204 127.648 106.641 262144 4298.840 134.339 191.774 94.105 121.699 126.223 104.826 524288 4239.543 132.486 196.344 89.397 118.316 124.292 103.460 1048576 4163.100 130.097 197.992 85.484 112.077 116.502 95.538 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-16*2fix : 0.055 0.868 10.365 79.530 191.311 208.973 -> 78.491 -> 2511.717 MByte/s p01 ring-8*4fix : 0.054 0.843 11.446 98.235 190.714 197.423 -> 80.989 -> 2591.638 MByte/s p02 ring-4*8fix : 0.053 0.824 11.467 98.757 153.422 190.005 -> 73.264 -> 2344.456 MByte/s p03 ring-2*16fix : 0.049 0.786 11.312 91.644 146.816 171.532 -> 69.308 -> 2217.841 MByte/s p04 ring-1*32fix : 0.048 0.779 11.317 97.242 198.840 212.248 -> 84.154 -> 2692.933 MByte/s p05 ring-1*32fix : 0.048 0.775 11.342 97.745 198.683 211.079 -> 83.986 -> 2687.541 MByte/s p06 random-cyc-1dim : 0.046 0.768 11.186 85.375 108.170 112.009 -> 52.629 -> 1684.127 MByte/s p07 random-cyc-1dim : 0.045 0.785 11.411 69.965 79.669 66.627 -> 38.816 -> 1242.102 MByte/s p08 random-cyc-1dim : 0.045 0.773 11.300 69.499 79.457 75.478 -> 38.696 -> 1238.273 MByte/s p09 random-cyc-1dim : 0.046 0.778 11.402 73.372 87.742 74.095 -> 41.201 -> 1318.444 MByte/s p10 random-cyc-1dim : 0.047 0.767 11.184 85.802 110.987 98.941 -> 51.527 -> 1648.854 MByte/s p11 random-cyc-1dim : 0.046 0.774 11.267 73.373 103.552 103.167 -> 48.124 -> 1539.969 MByte/s p12 random-cyc-1dim : 0.046 0.774 11.247 82.985 103.340 98.513 -> 49.849 -> 1595.161 MByte/s p13 random-cyc-1dim : 0.045 0.769 11.144 80.551 97.464 87.636 -> 45.937 -> 1469.983 MByte/s p14 random-cyc-1dim : 0.044 0.761 11.085 63.077 70.881 65.146 -> 34.845 -> 1115.050 MByte/s p15 random-cyc-1dim : 0.046 0.764 11.172 84.112 110.040 86.970 -> 50.642 -> 1620.559 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.044 0.763 11.157 65.194 72.209 60.792 -> 36.351 -> 1163.233 MByte/s p17 best bi-section : 0.030 0.505 7.567 78.645 190.668 209.202 -> 78.041 -> 2497.321 MByte/s p18 worst bi-section : 0.028 0.491 7.362 55.233 66.589 87.341 -> 33.805 -> 1081.760 MByte/s p19 acyclic-1dim-all : 0.047 0.774 11.227 97.616 165.720 141.603 -> 70.544 -> 2257.418 MByte/s p20 acyclic-2dim-all : 0.051 0.915 12.690 83.178 109.727 115.259 -> 53.146 -> 1700.687 MByte/s p21 acyclic-3dim-all : 0.049 0.873 12.131 81.184 109.811 131.384 -> 54.058 -> 1729.846 MByte/s p22 cyclic-1dim-all : 0.049 0.792 10.891 95.652 197.437 212.852 -> 83.372 -> 2667.895 MByte/s p23 cyclic-2dim-all : 0.058 1.049 14.604 86.745 118.184 120.172 -> 56.145 -> 1796.646 MByte/s p24 cyclic-3dim-all : 0.063 1.138 15.665 89.749 117.789 120.262 -> 57.956 -> 1854.601 MByte/s log_avg of all rings : 0.051 0.812 11.201 93.592 178.611 197.992 || 78.169 -> 2501.399 MByte/s log_avg of all random : 0.046 0.771 11.239 76.428 94.067 85.484 || 44.809 -> 1433.897 MByte/s log_avg(ring,random) : 0.048 0.791 11.220 84.576 129.620 130.097 || 59.183 -> 1893.872 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 1893.872 MByte/s on 32 processes ( = 59.183 MByte/s * 32 processes) system parameters : 32 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 = 1893.872 MB/s = 59.183 * 32 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E