b_eff = 5620.345 MB/s = 43.909 * 128 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 = 128 1-dim-paterns: size = 128 2-dim-paterns: size = 16 * 8 3-dim-paterns: size = 8 * 4 * 4 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 = 185.463 sec sum of max elapsed time per entries above = 191.563 sec difference = -6.099 sec = 3.3% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-64*2fix => 1 sendrecv_calls with 128 messages, i.e. msgs/used node, all nodes are used p01 ring-32*4fix => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p02 ring-16*8fix => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p03 ring-4*32fix => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p04 ring-2*64fix => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p05 ring-1*128fix => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 128 messages, i.e. msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 128 messages, i.e. msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 254 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 464 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 608 messages, i.e. msgs/used node, all nodes are used p22 cyclic-1dim-all => 2 sendrecv_calls with 256 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 512 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 6 sendrecv_calls with 768 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-64*2fix : 71.808 62.694 58.389 -> 71.808 -> 9191.420 MByte/s p01 ring-32*4fix : 58.015 67.937 44.063 -> 67.937 -> 8695.907 MByte/s p02 ring-16*8fix : 52.515 60.854 43.550 -> 60.854 -> 7789.258 MByte/s p03 ring-4*32fix : 54.105 68.733 45.976 -> 68.733 -> 8797.863 MByte/s p04 ring-2*64fix : 53.228 64.429 45.267 -> 64.429 -> 8246.953 MByte/s p05 ring-1*128fix : 53.815 62.966 44.187 -> 62.966 -> 8059.662 MByte/s p06 random-cyc-1dim : 25.943 25.286 23.574 -> 25.943 -> 3320.723 MByte/s p07 random-cyc-1dim : 29.717 26.326 25.583 -> 29.717 -> 3803.772 MByte/s p08 random-cyc-1dim : 29.947 28.977 26.912 -> 29.947 -> 3833.243 MByte/s p09 random-cyc-1dim : 25.298 25.243 23.101 -> 25.298 -> 3238.171 MByte/s p10 random-cyc-1dim : 29.620 28.321 26.357 -> 29.620 -> 3791.390 MByte/s p11 random-cyc-1dim : 25.388 24.084 22.834 -> 25.388 -> 3249.608 MByte/s p12 random-cyc-1dim : 32.167 30.745 28.074 -> 32.167 -> 4117.391 MByte/s p13 random-cyc-1dim : 27.203 24.449 24.138 -> 27.203 -> 3481.997 MByte/s p14 random-cyc-1dim : 24.253 22.020 21.420 -> 24.253 -> 3104.395 MByte/s p15 random-cyc-1dim : 25.654 23.525 22.808 -> 25.654 -> 3283.741 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 14.643 13.637 12.536 -> 14.643 -> 1874.342 MByte/s p17 best bi-section : 50.280 63.129 48.436 -> 63.129 -> 8080.476 MByte/s p18 worst bi-section : 14.541 16.117 14.074 -> 16.117 -> 2062.936 MByte/s p19 acyclic-1dim-all : 53.899 60.478 44.846 -> 60.478 -> 7741.243 MByte/s p20 acyclic-2dim-all : 43.007 47.899 37.502 -> 47.899 -> 6131.125 MByte/s p21 acyclic-3dim-all : 37.117 38.456 30.925 -> 38.456 -> 4922.317 MByte/s p22 cyclic-1dim-all : 53.502 63.776 44.085 -> 63.776 -> 8163.319 MByte/s p23 cyclic-2dim-all : 43.840 49.083 39.731 -> 49.083 -> 6282.624 MByte/s p24 cyclic-3dim-all : 37.054 38.876 31.828 -> 38.876 -> 4976.122 MByte/s log_avg of all rings : 56.891 64.540 46.648 || 66.017 -> 8450.133 MByte/s log_avg of all random : 27.407 25.773 24.397 || 27.407 -> 3508.116 MByte/s log_avg(ring,random) : 39.487 40.785 33.735 ||( 42.536 -> 5444.634)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-64*2fix : 70.981 71.075 71.340 -> 71.340 -> 9131.533 MByte/s p01 ring-32*4fix : 68.877 67.691 68.881 -> 68.881 -> 8816.730 MByte/s p02 ring-16*8fix : 60.715 61.741 61.445 -> 61.741 -> 7902.816 MByte/s p03 ring-4*32fix : 69.032 68.463 67.983 -> 69.032 -> 8836.104 MByte/s p04 ring-2*64fix : 63.890 63.735 64.559 -> 64.559 -> 8263.506 MByte/s p05 ring-1*128fix : 63.303 61.918 63.006 -> 63.303 -> 8102.843 MByte/s p06 random-cyc-1dim : 26.322 26.865 26.906 -> 26.906 -> 3443.964 MByte/s p07 random-cyc-1dim : 30.393 30.604 30.454 -> 30.604 -> 3917.357 MByte/s p08 random-cyc-1dim : 30.774 31.256 30.946 -> 31.256 -> 4000.733 MByte/s p09 random-cyc-1dim : 26.200 26.166 26.488 -> 26.488 -> 3390.424 MByte/s p10 random-cyc-1dim : 30.085 30.233 30.294 -> 30.294 -> 3877.617 MByte/s p11 random-cyc-1dim : 26.276 26.239 26.001 -> 26.276 -> 3363.312 MByte/s p12 random-cyc-1dim : 33.001 33.719 32.585 -> 33.719 -> 4316.063 MByte/s p13 random-cyc-1dim : 28.244 27.384 27.757 -> 28.244 -> 3615.187 MByte/s p14 random-cyc-1dim : 24.751 24.890 24.534 -> 24.890 -> 3185.888 MByte/s p15 random-cyc-1dim : 26.422 26.351 26.678 -> 26.678 -> 3414.789 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 14.729 14.953 14.785 -> 14.953 -> 1914.019 MByte/s p17 best bi-section : 63.912 62.984 62.950 -> 63.912 -> 8180.679 MByte/s p18 worst bi-section : 16.418 16.481 16.317 -> 16.481 -> 2109.593 MByte/s p19 acyclic-1dim-all : 59.633 59.982 59.902 -> 59.982 -> 7677.683 MByte/s p20 acyclic-2dim-all : 49.173 49.099 49.362 -> 49.362 -> 6318.343 MByte/s p21 acyclic-3dim-all : 43.210 43.193 43.018 -> 43.210 -> 5530.922 MByte/s p22 cyclic-1dim-all : 62.624 63.800 63.404 -> 63.800 -> 8166.390 MByte/s p23 cyclic-2dim-all : 48.397 49.499 49.324 -> 49.499 -> 6335.865 MByte/s p24 cyclic-3dim-all : 42.203 42.002 42.244 -> 42.244 -> 5407.212 MByte/s log_avg of all rings : 66.029 65.677 66.111 || 66.385 -> 8497.333 MByte/s log_avg of all random : 28.135 28.244 28.157 || 28.415 -> 3637.132 MByte/s log_avg(ring,random) : 43.102 43.069 43.145 ||( 43.432 -> 5559.309)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-64*2fix p00 method 0 : 0.056 0.869 9.941 50.912 179.706 207.808 -> 71.808 -> 9191.420 MByte/s p00 method 1 : 0.010 0.164 2.578 36.004 162.556 206.559 -> 62.694 -> 8024.874 MByte/s p00 method 2 : 0.025 0.390 5.335 26.145 165.673 159.070 -> 58.389 -> 7473.750 MByte/s p01 ring-32*4fix p01 method 0 : 0.054 0.850 9.734 49.535 147.443 153.810 -> 58.015 -> 7425.959 MByte/s p01 method 1 : 0.018 0.299 4.607 56.609 173.250 194.662 -> 67.937 -> 8695.907 MByte/s p01 method 2 : 0.021 0.348 4.591 29.199 116.624 99.157 -> 44.063 -> 5640.023 MByte/s p02 ring-16*8fix p02 method 0 : 0.052 0.828 9.546 47.594 133.985 129.935 -> 52.515 -> 6721.978 MByte/s p02 method 1 : 0.018 0.299 4.603 56.049 152.833 183.260 -> 60.854 -> 7789.258 MByte/s p02 method 2 : 0.021 0.342 4.430 29.797 115.154 109.641 -> 43.550 -> 5574.384 MByte/s p03 ring-4*32fix p03 method 0 : 0.047 0.782 8.532 39.097 142.559 131.639 -> 54.105 -> 6925.384 MByte/s p03 method 1 : 0.018 0.299 4.598 55.909 170.449 204.384 -> 68.733 -> 8797.863 MByte/s p03 method 2 : 0.022 0.343 4.479 29.849 124.088 107.322 -> 45.976 -> 5884.951 MByte/s p04 ring-2*64fix p04 method 0 : 0.047 0.762 8.313 38.512 139.881 129.928 -> 53.228 -> 6813.195 MByte/s p04 method 1 : 0.018 0.299 4.585 55.555 163.720 187.219 -> 64.429 -> 8246.953 MByte/s p04 method 2 : 0.022 0.341 4.500 29.795 124.139 103.672 -> 45.267 -> 5794.187 MByte/s p05 ring-1*128fix p05 method 0 : 0.046 0.755 8.329 39.063 141.363 132.172 -> 53.815 -> 6888.286 MByte/s p05 method 1 : 0.018 0.299 4.582 56.766 142.670 191.822 -> 62.966 -> 8059.662 MByte/s p05 method 2 : 0.022 0.339 4.500 29.431 120.804 100.378 -> 44.187 -> 5655.977 MByte/s p06 random-cyc-1dim p06 method 0 : 0.043 0.648 7.570 34.753 57.640 49.753 -> 25.943 -> 3320.723 MByte/s p06 method 1 : 0.018 0.294 4.527 48.022 56.437 40.703 -> 25.286 -> 3236.606 MByte/s p06 method 2 : 0.021 0.332 4.276 28.012 54.621 43.252 -> 23.574 -> 3017.521 MByte/s p07 random-cyc-1dim p07 method 0 : 0.043 0.603 7.513 35.712 68.435 58.788 -> 29.717 -> 3803.772 MByte/s p07 method 1 : 0.018 0.296 4.552 49.836 58.843 36.881 -> 26.326 -> 3369.777 MByte/s p07 method 2 : 0.021 0.330 4.270 28.300 61.222 54.300 -> 25.583 -> 3274.632 MByte/s p08 random-cyc-1dim p08 method 0 : 0.043 0.677 7.650 36.369 71.676 59.515 -> 29.947 -> 3833.243 MByte/s p08 method 1 : 0.018 0.297 4.578 51.383 66.897 47.206 -> 28.977 -> 3709.003 MByte/s p08 method 2 : 0.021 0.330 4.222 29.071 66.131 53.362 -> 26.912 -> 3444.682 MByte/s p09 random-cyc-1dim p09 method 0 : 0.042 0.691 7.528 34.517 57.167 46.556 -> 25.298 -> 3238.171 MByte/s p09 method 1 : 0.018 0.295 4.539 44.897 56.436 38.985 -> 25.243 -> 3231.054 MByte/s p09 method 2 : 0.020 0.331 4.269 27.778 53.948 50.395 -> 23.101 -> 2956.887 MByte/s p10 random-cyc-1dim p10 method 0 : 0.042 0.678 7.276 36.397 69.060 55.773 -> 29.620 -> 3791.390 MByte/s p10 method 1 : 0.018 0.298 4.579 50.732 64.257 48.056 -> 28.321 -> 3625.146 MByte/s p10 method 2 : 0.021 0.331 4.132 29.002 64.588 49.370 -> 26.357 -> 3373.641 MByte/s p11 random-cyc-1dim p11 method 0 : 0.042 0.669 7.407 34.989 57.581 41.208 -> 25.388 -> 3249.608 MByte/s p11 method 1 : 0.018 0.295 4.530 47.948 54.654 27.750 -> 24.084 -> 3082.745 MByte/s p11 method 2 : 0.020 0.330 4.209 28.615 54.443 41.485 -> 22.834 -> 2922.813 MByte/s p12 random-cyc-1dim p12 method 0 : 0.043 0.682 7.596 36.136 76.486 68.808 -> 32.167 -> 4117.391 MByte/s p12 method 1 : 0.018 0.295 4.552 51.375 70.638 52.943 -> 30.745 -> 3935.321 MByte/s p12 method 2 : 0.021 0.329 4.296 28.613 69.139 57.954 -> 28.074 -> 3593.524 MByte/s p13 random-cyc-1dim p13 method 0 : 0.043 0.671 7.565 35.882 63.657 42.791 -> 27.203 -> 3481.997 MByte/s p13 method 1 : 0.018 0.296 4.536 47.916 50.874 29.625 -> 24.449 -> 3129.434 MByte/s p13 method 2 : 0.020 0.328 4.174 28.636 57.571 46.158 -> 24.138 -> 3089.613 MByte/s p14 random-cyc-1dim p14 method 0 : 0.043 0.636 7.568 35.023 52.817 34.824 -> 24.253 -> 3104.395 MByte/s p14 method 1 : 0.018 0.297 4.542 48.206 46.195 26.354 -> 22.020 -> 2818.567 MByte/s p14 method 2 : 0.020 0.333 4.266 28.867 50.469 35.377 -> 21.420 -> 2741.740 MByte/s p15 random-cyc-1dim p15 method 0 : 0.043 0.688 7.468 34.954 59.018 45.314 -> 25.654 -> 3283.741 MByte/s p15 method 1 : 0.018 0.298 4.576 48.503 51.704 28.311 -> 23.525 -> 3011.161 MByte/s p15 method 2 : 0.021 0.330 4.208 28.419 54.157 43.719 -> 22.808 -> 2919.427 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.041 0.664 6.784 25.963 28.734 20.259 -> 14.643 -> 1874.342 MByte/s p16 method 1 : 0.018 0.293 4.513 29.811 26.759 17.450 -> 13.637 -> 1745.491 MByte/s p16 method 2 : 0.020 0.315 4.169 24.155 24.913 16.956 -> 12.536 -> 1604.660 MByte/s p17 best bi-section p17 method 0 : 0.033 0.499 5.488 27.698 128.131 157.185 -> 50.280 -> 6435.790 MByte/s p17 method 1 : 0.010 0.164 2.562 36.616 162.707 206.952 -> 63.129 -> 8080.476 MByte/s p17 method 2 : 0.015 0.233 3.317 24.314 125.402 152.779 -> 48.436 -> 6199.760 MByte/s p18 worst bi-section p18 method 0 : 0.023 0.370 4.418 23.220 29.353 27.192 -> 14.541 -> 1861.208 MByte/s p18 method 1 : 0.010 0.164 2.555 24.678 30.750 49.955 -> 16.117 -> 2062.936 MByte/s p18 method 2 : 0.013 0.211 2.902 20.933 29.759 33.327 -> 14.074 -> 1801.502 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.047 0.756 8.461 38.775 141.253 133.120 -> 53.899 -> 6899.035 MByte/s p19 method 1 : 0.018 0.297 4.571 57.121 160.374 142.953 -> 60.478 -> 7741.243 MByte/s p19 method 2 : 0.021 0.338 4.385 29.636 124.132 96.320 -> 44.846 -> 5740.335 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.037 0.567 6.402 29.360 110.609 111.258 -> 43.007 -> 5504.865 MByte/s p20 method 1 : 0.028 0.474 7.064 70.156 112.969 93.234 -> 47.899 -> 6131.125 MByte/s p20 method 2 : 0.020 0.312 3.962 27.012 99.959 79.686 -> 37.502 -> 4800.318 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.032 0.491 5.432 25.800 94.813 97.677 -> 37.117 -> 4750.994 MByte/s p21 method 1 : 0.032 0.559 8.072 66.305 81.265 73.803 -> 38.456 -> 4922.317 MByte/s p21 method 2 : 0.017 0.269 3.456 22.901 80.520 69.872 -> 30.925 -> 3958.455 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.047 0.755 8.052 38.000 140.025 130.005 -> 53.502 -> 6848.193 MByte/s p22 method 1 : 0.018 0.295 4.545 55.981 154.132 187.975 -> 63.776 -> 8163.319 MByte/s p22 method 2 : 0.022 0.331 4.436 29.750 117.263 104.056 -> 44.085 -> 5642.899 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.040 0.628 6.855 32.128 113.198 107.376 -> 43.840 -> 5611.564 MByte/s p23 method 1 : 0.029 0.497 7.433 70.547 113.147 106.611 -> 49.083 -> 6282.624 MByte/s p23 method 2 : 0.022 0.343 4.411 29.688 105.932 81.004 -> 39.731 -> 5085.548 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.037 0.573 6.314 30.944 92.369 89.523 -> 37.054 -> 4742.951 MByte/s p24 method 1 : 0.037 0.644 9.475 66.238 82.420 66.097 -> 38.876 -> 4976.122 MByte/s p24 method 2 : 0.021 0.329 4.214 28.313 83.397 63.490 -> 31.828 -> 4073.996 MByte/s log_avg of all rings - ring, method 0 : 0.050 0.807 9.040 43.799 146.799 145.244 || 56.891 -> 7282.095 MByte/s - ring, method 1 : 0.016 0.271 4.173 52.161 160.566 194.468 || 64.540 -> 8261.130 MByte/s - ring, method 2 : 0.022 0.350 4.629 29.005 126.718 111.595 || 46.648 -> 5970.947 MByte/s log_avg of all random - random, method 0 : 0.043 0.664 7.513 35.467 62.939 49.399 || 27.407 -> 3508.116 MByte/s - random, method 1 : 0.018 0.296 4.551 48.845 57.246 36.598 || 25.773 -> 3298.955 MByte/s - random, method 2 : 0.021 0.330 4.232 28.529 58.336 47.081 || 24.397 -> 3122.803 MByte/s log_avg(ring,random) - average, method 0 : 0.046 0.732 8.241 39.413 96.122 84.705 || 39.487 -> 5054.348 MByte/s - average, method 1 : 0.017 0.283 4.358 50.476 95.873 84.363 || 40.785 -> 5220.450 MByte/s - average, method 2 : 0.021 0.340 4.426 28.766 85.978 72.485 || 33.735 -> 4318.112 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 5.929 0.046 0.050 0.043 0.046 0.017 0.021 2 11.783 0.092 0.100 0.085 0.092 0.034 0.043 4 23.767 0.186 0.202 0.171 0.186 0.071 0.085 8 47.563 0.372 0.405 0.341 0.372 0.143 0.171 16 93.662 0.732 0.807 0.664 0.732 0.283 0.340 32 143.577 1.122 1.251 1.006 1.122 0.560 0.585 64 264.461 2.066 2.281 1.871 2.066 1.097 1.123 128 568.074 4.438 4.932 3.993 4.438 2.237 2.312 256 1054.891 8.241 9.040 7.513 8.241 4.358 4.426 512 1860.878 14.538 15.797 13.380 14.538 8.494 8.004 1024 2887.876 22.562 25.000 20.361 22.562 16.361 13.910 2048 4176.717 32.631 34.559 30.810 30.745 30.048 21.294 4096 6650.167 51.954 55.262 48.845 39.413 50.476 28.766 8192 9128.637 71.317 86.534 58.776 60.548 69.379 57.357 16384 10915.213 85.275 118.038 61.606 78.392 83.150 72.808 32768 12169.725 95.076 144.056 62.750 89.085 92.282 81.154 65536 12975.600 101.372 163.273 62.939 96.122 95.873 85.978 131072 13327.742 104.123 174.346 62.184 97.888 97.790 86.830 262144 13535.392 105.745 184.749 60.526 96.551 96.136 84.142 524288 13107.497 102.402 191.149 54.859 90.953 92.893 79.254 1048576 12663.853 98.936 194.663 50.284 84.705 84.363 72.485 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-64*2fix : 0.056 0.869 9.941 50.912 179.706 207.808 -> 71.808 -> 9191.420 MByte/s p01 ring-32*4fix : 0.054 0.850 9.734 56.609 173.250 194.662 -> 69.633 -> 8913.079 MByte/s p02 ring-16*8fix : 0.052 0.828 9.546 56.049 152.833 183.260 -> 62.373 -> 7983.805 MByte/s p03 ring-4*32fix : 0.047 0.782 8.532 55.909 170.449 204.384 -> 69.703 -> 8921.930 MByte/s p04 ring-2*64fix : 0.047 0.762 8.313 55.555 163.720 187.219 -> 65.358 -> 8365.836 MByte/s p05 ring-1*128fix : 0.046 0.755 8.329 56.766 142.670 191.822 -> 63.849 -> 8172.611 MByte/s p06 random-cyc-1dim : 0.043 0.648 7.570 48.022 57.640 49.753 -> 27.261 -> 3489.384 MByte/s p07 random-cyc-1dim : 0.043 0.603 7.513 49.836 68.435 58.788 -> 31.014 -> 3969.846 MByte/s p08 random-cyc-1dim : 0.043 0.677 7.650 51.383 71.676 59.515 -> 31.618 -> 4047.114 MByte/s p09 random-cyc-1dim : 0.042 0.691 7.528 44.897 57.167 50.395 -> 26.761 -> 3425.352 MByte/s p10 random-cyc-1dim : 0.042 0.678 7.276 50.732 69.060 55.773 -> 30.836 -> 3946.960 MByte/s p11 random-cyc-1dim : 0.042 0.669 7.407 47.948 57.581 41.485 -> 26.508 -> 3393.083 MByte/s p12 random-cyc-1dim : 0.043 0.682 7.596 51.375 76.486 68.808 -> 34.014 -> 4353.833 MByte/s p13 random-cyc-1dim : 0.043 0.671 7.565 47.916 63.657 46.158 -> 28.514 -> 3649.741 MByte/s p14 random-cyc-1dim : 0.043 0.636 7.568 48.206 52.817 35.377 -> 25.352 -> 3245.021 MByte/s p15 random-cyc-1dim : 0.043 0.688 7.468 48.503 59.018 45.314 -> 26.950 -> 3449.552 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.041 0.664 6.784 29.811 28.734 20.259 -> 15.105 -> 1933.434 MByte/s p17 best bi-section : 0.033 0.499 5.488 36.616 162.707 206.952 -> 64.045 -> 8197.727 MByte/s p18 worst bi-section : 0.023 0.370 4.418 24.678 30.750 49.955 -> 16.599 -> 2124.633 MByte/s p19 acyclic-1dim-all : 0.047 0.756 8.461 57.121 160.374 142.953 -> 61.394 -> 7858.458 MByte/s p20 acyclic-2dim-all : 0.037 0.567 7.064 70.156 112.969 111.258 -> 49.857 -> 6381.657 MByte/s p21 acyclic-3dim-all : 0.032 0.559 8.072 66.305 94.813 97.677 -> 43.465 -> 5563.462 MByte/s p22 cyclic-1dim-all : 0.047 0.755 8.052 55.981 154.132 187.975 -> 64.691 -> 8280.479 MByte/s p23 cyclic-2dim-all : 0.040 0.628 7.433 70.547 113.198 107.376 -> 49.860 -> 6382.138 MByte/s p24 cyclic-3dim-all : 0.037 0.644 9.475 66.238 92.369 89.523 -> 42.489 -> 5438.632 MByte/s log_avg of all rings : 0.050 0.807 9.040 55.262 163.273 194.663 || 67.032 -> 8580.084 MByte/s log_avg of all random : 0.043 0.664 7.513 48.845 62.939 50.284 || 28.762 -> 3681.582 MByte/s log_avg(ring,random) : 0.046 0.732 8.241 51.954 101.372 98.936 || 43.909 -> 5620.345 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 5620.345 MByte/s on 128 processes ( = 43.909 MByte/s * 128 processes) system parameters : 128 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 = 5620.345 MB/s = 43.909 * 128 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E