b_eff = 3170.223 MB/s = 46.621 * 68 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 = 68 1-dim-paterns: size = 68 2-dim-paterns: size = 17 * 4 3-dim-paterns: size = 17 * 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 = 159.127 sec sum of max elapsed time per entries above = 163.704 sec difference = -4.577 sec = 2.9% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-34*2fix => 1 sendrecv_calls with 68 messages, i.e. msgs/used node, all nodes are used p01 ring-17*4fix => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p02 ring-8*8&+1 => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p03 ring-4*17fix => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p04 ring-2*34fix => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p05 ring-1*68fix => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 68 messages, i.e. msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 68 messages, i.e. msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 134 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 230 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 264 messages, i.e. msgs/used node, all nodes are used p22 cyclic-1dim-all => 2 sendrecv_calls with 136 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 272 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 4 sendrecv_calls with 272 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-34*2fix : 72.066 68.964 59.912 -> 72.066 -> 4900.508 MByte/s p01 ring-17*4fix : 59.786 74.037 45.167 -> 74.037 -> 5034.527 MByte/s p02 ring-8*8&+1 : 53.889 66.444 44.759 -> 66.444 -> 4518.160 MByte/s p03 ring-4*17fix : 53.685 69.884 45.180 -> 69.884 -> 4752.090 MByte/s p04 ring-2*34fix : 54.271 72.535 45.466 -> 72.535 -> 4932.405 MByte/s p05 ring-1*68fix : 53.526 62.984 45.144 -> 62.984 -> 4282.921 MByte/s p06 random-cyc-1dim : 29.317 31.940 27.148 -> 31.940 -> 2171.942 MByte/s p07 random-cyc-1dim : 30.691 29.852 26.343 -> 30.691 -> 2086.987 MByte/s p08 random-cyc-1dim : 25.727 26.188 23.601 -> 26.188 -> 1780.800 MByte/s p09 random-cyc-1dim : 27.987 28.763 24.992 -> 28.763 -> 1955.865 MByte/s p10 random-cyc-1dim : 31.467 32.262 26.961 -> 32.262 -> 2193.806 MByte/s p11 random-cyc-1dim : 27.516 27.130 24.550 -> 27.516 -> 1871.088 MByte/s p12 random-cyc-1dim : 32.386 30.238 28.156 -> 32.386 -> 2202.265 MByte/s p13 random-cyc-1dim : 23.097 23.547 20.945 -> 23.547 -> 1601.188 MByte/s p14 random-cyc-1dim : 27.036 27.363 24.248 -> 27.363 -> 1860.679 MByte/s p15 random-cyc-1dim : 29.727 30.053 27.015 -> 30.053 -> 2043.632 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 19.855 19.794 17.287 -> 19.855 -> 1350.132 MByte/s p17 best bi-section : 51.204 68.949 48.846 -> 68.949 -> 4688.540 MByte/s p18 worst bi-section : 14.618 17.647 14.294 -> 17.647 -> 1199.966 MByte/s p19 acyclic-1dim-all : 54.517 65.669 45.209 -> 65.669 -> 4465.470 MByte/s p20 acyclic-2dim-all : 41.970 46.571 35.274 -> 46.571 -> 3166.849 MByte/s p21 acyclic-3dim-all : 42.589 48.721 38.121 -> 48.721 -> 3313.001 MByte/s p22 cyclic-1dim-all : 53.201 63.311 45.414 -> 63.311 -> 4305.153 MByte/s p23 cyclic-2dim-all : 46.036 47.759 37.528 -> 47.759 -> 3247.594 MByte/s p24 cyclic-3dim-all : 45.663 52.142 39.734 -> 52.142 -> 3545.676 MByte/s log_avg of all rings : 57.518 69.042 47.323 || 69.550 -> 4729.416 MByte/s log_avg of all random : 28.366 28.615 25.310 || 28.933 -> 1967.433 MByte/s log_avg(ring,random) : 40.392 44.448 34.608 ||( 44.859 -> 3050.379)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-34*2fix : 71.884 71.861 71.846 -> 71.884 -> 4888.112 MByte/s p01 ring-17*4fix : 72.447 72.837 74.322 -> 74.322 -> 5053.863 MByte/s p02 ring-8*8&+1 : 66.380 65.470 65.844 -> 66.380 -> 4513.821 MByte/s p03 ring-4*17fix : 69.677 69.499 68.990 -> 69.677 -> 4738.046 MByte/s p04 ring-2*34fix : 71.226 71.732 71.490 -> 71.732 -> 4877.791 MByte/s p05 ring-1*68fix : 62.797 62.843 62.702 -> 62.843 -> 4273.350 MByte/s p06 random-cyc-1dim : 31.761 32.252 32.647 -> 32.647 -> 2220.023 MByte/s p07 random-cyc-1dim : 32.394 32.771 32.221 -> 32.771 -> 2228.448 MByte/s p08 random-cyc-1dim : 28.412 27.781 27.722 -> 28.412 -> 1932.025 MByte/s p09 random-cyc-1dim : 29.831 29.250 30.086 -> 30.086 -> 2045.854 MByte/s p10 random-cyc-1dim : 33.985 34.013 34.023 -> 34.023 -> 2313.567 MByte/s p11 random-cyc-1dim : 28.696 28.995 29.014 -> 29.014 -> 1972.975 MByte/s p12 random-cyc-1dim : 33.856 34.596 34.160 -> 34.596 -> 2352.554 MByte/s p13 random-cyc-1dim : 24.867 24.759 24.626 -> 24.867 -> 1690.987 MByte/s p14 random-cyc-1dim : 28.500 28.862 28.757 -> 28.862 -> 1962.584 MByte/s p15 random-cyc-1dim : 32.248 32.630 31.788 -> 32.630 -> 2218.844 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 20.292 20.631 20.721 -> 20.721 -> 1409.050 MByte/s p17 best bi-section : 68.909 67.963 69.102 -> 69.102 -> 4698.946 MByte/s p18 worst bi-section : 17.653 17.556 17.628 -> 17.653 -> 1200.390 MByte/s p19 acyclic-1dim-all : 64.085 65.175 64.784 -> 65.175 -> 4431.924 MByte/s p20 acyclic-2dim-all : 49.662 49.275 49.781 -> 49.781 -> 3385.125 MByte/s p21 acyclic-3dim-all : 51.322 51.610 51.755 -> 51.755 -> 3519.321 MByte/s p22 cyclic-1dim-all : 63.379 62.004 62.187 -> 63.379 -> 4309.772 MByte/s p23 cyclic-2dim-all : 52.748 52.201 52.891 -> 52.891 -> 3596.562 MByte/s p24 cyclic-3dim-all : 53.024 52.988 53.053 -> 53.053 -> 3607.585 MByte/s log_avg of all rings : 68.980 68.940 69.086 || 69.365 -> 4716.827 MByte/s log_avg of all random : 30.327 30.441 30.363 || 30.649 -> 2084.148 MByte/s log_avg(ring,random) : 45.738 45.811 45.800 ||( 46.108 -> 3135.373)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-34*2fix p00 method 0 : 0.056 0.871 10.377 51.138 179.585 206.991 -> 72.066 -> 4900.508 MByte/s p00 method 1 : 0.015 0.264 4.073 51.220 176.472 207.264 -> 68.964 -> 4689.529 MByte/s p00 method 2 : 0.025 0.399 5.308 26.144 168.137 162.584 -> 59.912 -> 4074.034 MByte/s p01 ring-17*4fix p01 method 0 : 0.054 0.843 9.773 50.039 153.330 153.833 -> 59.786 -> 4065.428 MByte/s p01 method 1 : 0.027 0.458 6.943 75.041 182.932 194.916 -> 74.037 -> 5034.527 MByte/s p01 method 2 : 0.021 0.348 4.632 29.365 119.093 109.800 -> 45.167 -> 3071.350 MByte/s p02 ring-8*8&+1 p02 method 0 : 0.050 0.818 9.515 47.039 135.962 135.656 -> 53.889 -> 3664.460 MByte/s p02 method 1 : 0.027 0.457 6.926 73.284 152.796 186.666 -> 66.444 -> 4518.160 MByte/s p02 method 2 : 0.022 0.346 4.459 30.024 117.683 114.134 -> 44.759 -> 3043.624 MByte/s p03 ring-4*17fix p03 method 0 : 0.048 0.786 8.628 44.521 136.671 132.636 -> 53.685 -> 3650.551 MByte/s p03 method 1 : 0.027 0.454 6.856 74.105 163.682 194.701 -> 69.884 -> 4752.090 MByte/s p03 method 2 : 0.021 0.344 4.486 30.088 119.966 104.888 -> 45.180 -> 3072.238 MByte/s p04 ring-2*34fix p04 method 0 : 0.048 0.783 8.549 40.710 141.366 130.595 -> 54.271 -> 3690.440 MByte/s p04 method 1 : 0.027 0.457 6.916 72.574 176.811 197.023 -> 72.535 -> 4932.405 MByte/s p04 method 2 : 0.022 0.344 4.507 29.942 124.962 101.655 -> 45.466 -> 3091.678 MByte/s p05 ring-1*68fix p05 method 0 : 0.048 0.763 8.302 38.743 139.182 135.402 -> 53.526 -> 3639.761 MByte/s p05 method 1 : 0.026 0.452 6.823 69.673 150.117 163.606 -> 62.984 -> 4282.921 MByte/s p05 method 2 : 0.022 0.343 4.479 29.808 119.482 105.822 -> 45.144 -> 3069.799 MByte/s p06 random-cyc-1dim p06 method 0 : 0.045 0.692 7.692 36.533 70.437 47.201 -> 29.317 -> 1993.539 MByte/s p06 method 1 : 0.026 0.447 6.751 59.085 69.104 52.315 -> 31.940 -> 2171.942 MByte/s p06 method 2 : 0.021 0.335 4.334 29.000 65.354 57.991 -> 27.148 -> 1846.081 MByte/s p07 random-cyc-1dim p07 method 0 : 0.044 0.700 7.674 36.424 69.887 66.832 -> 30.691 -> 2086.987 MByte/s p07 method 1 : 0.026 0.450 6.794 58.902 61.412 49.326 -> 29.852 -> 2029.970 MByte/s p07 method 2 : 0.021 0.334 4.324 28.699 63.692 54.464 -> 26.343 -> 1791.302 MByte/s p08 random-cyc-1dim p08 method 0 : 0.043 0.687 7.540 35.421 59.275 42.339 -> 25.727 -> 1749.413 MByte/s p08 method 1 : 0.026 0.445 6.702 53.906 55.414 31.716 -> 26.188 -> 1780.800 MByte/s p08 method 2 : 0.021 0.328 4.238 28.862 54.201 55.283 -> 23.601 -> 1604.838 MByte/s p09 random-cyc-1dim p09 method 0 : 0.044 0.707 7.637 35.789 64.404 51.196 -> 27.987 -> 1903.106 MByte/s p09 method 1 : 0.026 0.448 6.760 55.686 60.853 43.693 -> 28.763 -> 1955.865 MByte/s p09 method 2 : 0.021 0.335 4.348 29.046 59.466 50.211 -> 24.992 -> 1699.474 MByte/s p10 random-cyc-1dim p10 method 0 : 0.044 0.707 7.638 36.424 78.281 49.062 -> 31.467 -> 2139.726 MByte/s p10 method 1 : 0.027 0.451 6.848 62.358 70.098 41.831 -> 32.262 -> 2193.806 MByte/s p10 method 2 : 0.021 0.335 4.345 29.270 64.383 56.876 -> 26.961 -> 1833.324 MByte/s p11 random-cyc-1dim p11 method 0 : 0.043 0.690 7.622 35.768 61.473 48.727 -> 27.516 -> 1871.088 MByte/s p11 method 1 : 0.026 0.449 6.753 51.685 57.206 36.596 -> 27.130 -> 1844.842 MByte/s p11 method 2 : 0.021 0.335 4.258 28.789 57.524 47.469 -> 24.550 -> 1669.389 MByte/s p12 random-cyc-1dim p12 method 0 : 0.044 0.687 7.733 37.447 76.110 65.302 -> 32.386 -> 2202.265 MByte/s p12 method 1 : 0.027 0.451 6.815 60.980 68.688 37.655 -> 30.238 -> 2056.179 MByte/s p12 method 2 : 0.021 0.331 4.314 29.249 70.149 63.429 -> 28.156 -> 1914.603 MByte/s p13 random-cyc-1dim p13 method 0 : 0.044 0.711 7.640 34.395 53.702 34.229 -> 23.097 -> 1570.592 MByte/s p13 method 1 : 0.027 0.450 6.807 50.576 49.574 26.834 -> 23.547 -> 1601.188 MByte/s p13 method 2 : 0.021 0.332 4.340 28.591 50.352 32.113 -> 20.945 -> 1424.264 MByte/s p14 random-cyc-1dim p14 method 0 : 0.044 0.693 7.680 35.644 63.418 43.147 -> 27.036 -> 1838.462 MByte/s p14 method 1 : 0.027 0.448 6.771 53.582 55.472 43.074 -> 27.363 -> 1860.679 MByte/s p14 method 2 : 0.021 0.334 4.289 28.888 56.812 52.484 -> 24.248 -> 1648.883 MByte/s p15 random-cyc-1dim p15 method 0 : 0.043 0.703 7.727 36.427 69.432 48.561 -> 29.727 -> 2021.444 MByte/s p15 method 1 : 0.027 0.450 6.806 60.490 67.158 33.657 -> 30.053 -> 2043.632 MByte/s p15 method 2 : 0.021 0.331 4.324 29.128 62.649 53.298 -> 27.015 -> 1837.025 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.042 0.696 7.740 30.285 39.211 44.412 -> 19.855 -> 1350.132 MByte/s p16 method 1 : 0.026 0.453 6.834 36.926 38.666 32.741 -> 19.794 -> 1345.995 MByte/s p16 method 2 : 0.020 0.331 4.240 27.006 36.911 32.462 -> 17.287 -> 1175.500 MByte/s p17 best bi-section p17 method 0 : 0.033 0.498 5.505 27.850 131.540 162.617 -> 51.204 -> 3481.866 MByte/s p17 method 1 : 0.016 0.262 4.057 50.864 176.179 208.265 -> 68.949 -> 4688.540 MByte/s p17 method 2 : 0.015 0.241 3.311 27.500 126.829 160.533 -> 48.846 -> 3321.515 MByte/s p18 worst bi-section p18 method 0 : 0.023 0.372 4.526 23.186 29.700 31.243 -> 14.618 -> 994.010 MByte/s p18 method 1 : 0.015 0.257 3.973 27.823 32.607 44.211 -> 17.647 -> 1199.966 MByte/s p18 method 2 : 0.015 0.242 3.323 22.166 30.228 34.275 -> 14.294 -> 971.970 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.047 0.759 8.349 37.991 141.174 142.574 -> 54.517 -> 3707.190 MByte/s p19 method 1 : 0.027 0.451 6.841 75.275 164.386 146.003 -> 65.669 -> 4465.470 MByte/s p19 method 2 : 0.021 0.338 4.436 29.033 123.761 104.328 -> 45.209 -> 3074.242 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.035 0.553 6.098 27.588 110.399 106.591 -> 41.970 -> 2853.954 MByte/s p20 method 1 : 0.036 0.629 9.106 76.273 106.437 79.043 -> 46.571 -> 3166.849 MByte/s p20 method 2 : 0.018 0.289 3.773 25.735 93.534 75.368 -> 35.274 -> 2398.614 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.032 0.498 5.731 27.833 109.565 117.538 -> 42.589 -> 2896.080 MByte/s p21 method 1 : 0.041 0.719 10.263 78.512 108.417 87.240 -> 48.721 -> 3313.001 MByte/s p21 method 2 : 0.017 0.269 3.573 26.244 101.069 93.744 -> 38.121 -> 2592.249 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.047 0.763 8.243 39.343 137.890 131.399 -> 53.201 -> 3617.694 MByte/s p22 method 1 : 0.026 0.442 6.672 68.571 151.397 166.080 -> 63.311 -> 4305.153 MByte/s p22 method 2 : 0.022 0.344 4.417 30.051 122.140 101.008 -> 45.414 -> 3088.159 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.040 0.634 6.914 32.679 117.179 122.774 -> 46.036 -> 3130.425 MByte/s p23 method 1 : 0.040 0.704 10.225 78.146 106.173 80.071 -> 47.759 -> 3247.594 MByte/s p23 method 2 : 0.021 0.344 4.323 30.056 98.881 76.463 -> 37.528 -> 2551.916 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.039 0.626 6.820 32.749 116.739 120.296 -> 45.663 -> 3105.111 MByte/s p24 method 1 : 0.041 0.705 10.221 78.273 112.243 116.769 -> 52.142 -> 3545.676 MByte/s p24 method 2 : 0.022 0.341 4.321 29.754 105.035 96.137 -> 39.734 -> 2701.894 MByte/s log_avg of all rings - ring, method 0 : 0.050 0.810 9.160 45.133 146.945 147.099 || 57.518 -> 3911.210 MByte/s - ring, method 1 : 0.024 0.416 6.314 68.745 166.662 190.190 || 69.042 -> 4694.855 MByte/s - ring, method 2 : 0.022 0.354 4.636 29.193 127.126 114.883 || 47.323 -> 3217.985 MByte/s log_avg of all random - random, method 0 : 0.044 0.698 7.658 36.019 66.245 48.798 || 28.366 -> 1928.857 MByte/s - random, method 1 : 0.026 0.449 6.781 56.586 61.124 38.941 || 28.615 -> 1945.820 MByte/s - random, method 2 : 0.021 0.333 4.311 28.952 60.195 51.640 || 25.310 -> 1721.051 MByte/s log_avg(ring,random) - average, method 0 : 0.047 0.752 8.376 40.319 98.663 84.724 || 40.392 -> 2746.665 MByte/s - average, method 1 : 0.025 0.432 6.543 62.370 100.931 86.059 || 44.448 -> 3022.473 MByte/s - average, method 2 : 0.021 0.343 4.471 29.072 87.478 77.023 || 34.608 -> 2353.362 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 3.195 0.047 0.050 0.044 0.047 0.025 0.021 2 6.395 0.094 0.101 0.088 0.094 0.051 0.043 4 12.821 0.189 0.203 0.175 0.189 0.110 0.086 8 25.671 0.378 0.407 0.350 0.378 0.220 0.172 16 51.115 0.752 0.810 0.698 0.752 0.432 0.343 32 77.548 1.140 1.266 1.027 1.140 0.852 0.593 64 141.850 2.086 2.290 1.900 2.086 1.650 1.133 128 310.768 4.570 5.051 4.135 4.570 3.398 2.352 256 569.542 8.376 9.160 7.658 8.376 6.543 4.471 512 1004.008 14.765 16.084 13.554 14.755 12.553 8.105 1024 1713.659 25.201 26.119 24.315 22.867 23.634 13.990 2048 2862.859 42.101 43.237 40.995 31.468 41.147 21.591 4096 4241.157 62.370 68.745 56.586 40.319 62.370 29.072 8192 5437.468 79.963 101.658 62.897 60.965 79.535 57.855 16384 6269.515 92.199 132.850 63.986 79.798 91.506 73.700 32768 6834.476 100.507 155.326 65.035 91.424 98.481 82.911 65536 7155.440 105.227 167.148 66.245 98.663 100.931 87.478 131072 7243.628 106.524 175.584 64.627 100.288 101.599 89.502 262144 7313.358 107.549 181.662 63.672 100.118 99.414 86.533 524288 7238.734 106.452 187.498 60.438 95.110 95.120 82.659 1048576 6855.645 100.818 190.190 53.443 84.724 86.059 77.023 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-34*2fix : 0.056 0.871 10.377 51.220 179.585 207.264 -> 72.105 -> 4903.111 MByte/s p01 ring-17*4fix : 0.054 0.843 9.773 75.041 182.932 194.916 -> 74.656 -> 5076.628 MByte/s p02 ring-8*8&+1 : 0.050 0.818 9.515 73.284 152.796 186.666 -> 66.963 -> 4553.488 MByte/s p03 ring-4*17fix : 0.048 0.786 8.628 74.105 163.682 194.701 -> 70.195 -> 4773.253 MByte/s p04 ring-2*34fix : 0.048 0.783 8.549 72.574 176.811 197.023 -> 72.833 -> 4952.655 MByte/s p05 ring-1*68fix : 0.048 0.763 8.302 69.673 150.117 163.606 -> 63.249 -> 4300.936 MByte/s p06 random-cyc-1dim : 0.045 0.692 7.692 59.085 70.437 57.991 -> 32.891 -> 2236.604 MByte/s p07 random-cyc-1dim : 0.044 0.700 7.674 58.902 69.887 66.832 -> 33.261 -> 2261.781 MByte/s p08 random-cyc-1dim : 0.043 0.687 7.540 53.906 59.275 55.283 -> 28.691 -> 1951.014 MByte/s p09 random-cyc-1dim : 0.044 0.707 7.637 55.686 64.404 51.196 -> 30.365 -> 2064.825 MByte/s p10 random-cyc-1dim : 0.044 0.707 7.638 62.358 78.281 56.876 -> 35.004 -> 2380.276 MByte/s p11 random-cyc-1dim : 0.043 0.690 7.622 51.685 61.473 48.727 -> 29.381 -> 1997.939 MByte/s p12 random-cyc-1dim : 0.044 0.687 7.733 60.980 76.110 65.302 -> 35.291 -> 2399.778 MByte/s p13 random-cyc-1dim : 0.044 0.711 7.640 50.576 53.702 34.229 -> 25.121 -> 1708.223 MByte/s p14 random-cyc-1dim : 0.044 0.693 7.680 53.582 63.418 52.484 -> 29.490 -> 2005.343 MByte/s p15 random-cyc-1dim : 0.043 0.703 7.727 60.490 69.432 53.298 -> 33.010 -> 2244.681 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.042 0.696 7.740 36.926 39.211 44.412 -> 20.910 -> 1421.908 MByte/s p17 best bi-section : 0.033 0.498 5.505 50.864 176.179 208.265 -> 69.213 -> 4706.457 MByte/s p18 worst bi-section : 0.023 0.372 4.526 27.823 32.607 44.211 -> 17.743 -> 1206.495 MByte/s p19 acyclic-1dim-all : 0.047 0.759 8.349 75.275 164.386 146.003 -> 65.926 -> 4482.952 MByte/s p20 acyclic-2dim-all : 0.036 0.629 9.106 76.273 110.399 106.591 -> 50.233 -> 3415.812 MByte/s p21 acyclic-3dim-all : 0.041 0.719 10.263 78.512 109.565 117.538 -> 52.042 -> 3538.823 MByte/s p22 cyclic-1dim-all : 0.047 0.763 8.243 68.571 151.397 166.080 -> 63.615 -> 4325.830 MByte/s p23 cyclic-2dim-all : 0.040 0.704 10.225 78.146 117.179 122.774 -> 53.389 -> 3630.438 MByte/s p24 cyclic-3dim-all : 0.041 0.705 10.221 78.273 116.739 120.296 -> 53.482 -> 3636.766 MByte/s log_avg of all rings : 0.050 0.810 9.160 68.745 167.148 190.190 || 69.892 -> 4752.635 MByte/s log_avg of all random : 0.044 0.698 7.658 56.586 66.245 53.443 || 31.098 -> 2114.683 MByte/s log_avg(ring,random) : 0.047 0.752 8.376 62.370 105.227 100.818 || 46.621 -> 3170.223 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 3170.223 MByte/s on 68 processes ( = 46.621 MByte/s * 68 processes) system parameters : 68 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 = 3170.223 MB/s = 46.621 * 68 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E