b_eff = 3269.166 MB/s = 50.295 * 65 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 = 65 1-dim-paterns: size = 65 2-dim-paterns: size = 13 * 5 3-dim-paterns: size = 4 * 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 = 165.245 sec sum of max elapsed time per entries above = 167.587 sec difference = -2.342 sec = 1.4% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-32*2&+1 => 1 sendrecv_calls with 65 messages, i.e. msgs/used node, all nodes are used p01 ring-16*4&+1 => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p02 ring-8*8&+1 => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p03 ring-4*16&+1 => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p04 ring-2*32&+1 => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p05 ring-1*65fix => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, 1 nodes are UNUSED p18 worst bi-section => 2 sendrecv_calls with 64 messages, i.e. msgs/used node, 1 nodes are UNUSED p19 acyclic-1dim-all => 2 sendrecv_calls with 128 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 224 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 288 messages, i.e. msgs/used node, 1 nodes are UNUSED p22 cyclic-1dim-all => 2 sendrecv_calls with 130 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 260 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 6 sendrecv_calls with 384 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-32*2&+1 : 68.781 69.624 52.987 -> 69.624 -> 4525.565 MByte/s p01 ring-16*4&+1 : 56.644 74.562 43.986 -> 74.562 -> 4846.528 MByte/s p02 ring-8*8&+1 : 53.388 66.445 45.327 -> 66.445 -> 4318.896 MByte/s p03 ring-4*16&+1 : 51.048 64.628 43.203 -> 64.628 -> 4200.796 MByte/s p04 ring-2*32&+1 : 54.473 77.080 46.390 -> 77.080 -> 5010.222 MByte/s p05 ring-1*65fix : 54.620 76.303 46.458 -> 76.303 -> 4959.671 MByte/s p06 random-cyc-1dim : 32.645 33.424 28.864 -> 33.424 -> 2172.590 MByte/s p07 random-cyc-1dim : 34.033 39.121 30.367 -> 39.121 -> 2542.864 MByte/s p08 random-cyc-1dim : 30.896 32.380 28.075 -> 32.380 -> 2104.730 MByte/s p09 random-cyc-1dim : 31.425 31.952 28.076 -> 31.952 -> 2076.905 MByte/s p10 random-cyc-1dim : 27.751 28.005 25.558 -> 28.005 -> 1820.322 MByte/s p11 random-cyc-1dim : 33.219 32.600 28.864 -> 33.219 -> 2159.261 MByte/s p12 random-cyc-1dim : 32.720 33.538 29.402 -> 33.538 -> 2179.950 MByte/s p13 random-cyc-1dim : 28.916 29.205 25.970 -> 29.205 -> 1898.356 MByte/s p14 random-cyc-1dim : 34.028 34.843 30.763 -> 34.843 -> 2264.781 MByte/s p15 random-cyc-1dim : 32.914 32.354 28.510 -> 32.914 -> 2139.382 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 20.898 21.619 19.585 -> 21.619 -> 1405.208 MByte/s p17 best bi-section : 50.768 68.602 47.984 -> 68.602 -> 4459.119 MByte/s p18 worst bi-section : 17.809 20.210 16.497 -> 20.210 -> 1313.667 MByte/s p19 acyclic-1dim-all : 54.219 65.373 45.740 -> 65.373 -> 4249.275 MByte/s p20 acyclic-2dim-all : 39.780 46.021 34.306 -> 46.021 -> 2991.346 MByte/s p21 acyclic-3dim-all : 35.862 38.683 31.089 -> 38.683 -> 2514.377 MByte/s p22 cyclic-1dim-all : 54.357 76.030 45.658 -> 76.030 -> 4941.974 MByte/s p23 cyclic-2dim-all : 40.978 45.849 35.695 -> 45.849 -> 2980.175 MByte/s p24 cyclic-3dim-all : 36.570 41.075 31.156 -> 41.075 -> 2669.849 MByte/s log_avg of all rings : 56.225 71.275 46.289 || 71.275 -> 4632.900 MByte/s log_avg of all random : 31.789 32.619 28.400 || 32.737 -> 2127.897 MByte/s log_avg(ring,random) : 42.277 48.218 36.257 ||( 48.305 -> 3139.799)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-32*2&+1 : 71.375 70.558 71.116 -> 71.375 -> 4639.369 MByte/s p01 ring-16*4&+1 : 73.568 72.674 74.060 -> 74.060 -> 4813.895 MByte/s p02 ring-8*8&+1 : 65.484 66.014 66.016 -> 66.016 -> 4291.030 MByte/s p03 ring-4*16&+1 : 63.631 62.214 63.516 -> 63.631 -> 4135.999 MByte/s p04 ring-2*32&+1 : 76.273 75.988 76.480 -> 76.480 -> 4971.230 MByte/s p05 ring-1*65fix : 75.232 75.412 75.698 -> 75.698 -> 4920.369 MByte/s p06 random-cyc-1dim : 35.661 35.073 34.819 -> 35.661 -> 2317.953 MByte/s p07 random-cyc-1dim : 38.623 38.946 38.520 -> 38.946 -> 2531.486 MByte/s p08 random-cyc-1dim : 33.287 33.282 33.323 -> 33.323 -> 2166.020 MByte/s p09 random-cyc-1dim : 34.418 34.033 33.939 -> 34.418 -> 2237.202 MByte/s p10 random-cyc-1dim : 29.910 29.960 29.877 -> 29.960 -> 1947.388 MByte/s p11 random-cyc-1dim : 36.360 35.504 35.699 -> 36.360 -> 2363.374 MByte/s p12 random-cyc-1dim : 35.523 35.376 35.665 -> 35.665 -> 2318.235 MByte/s p13 random-cyc-1dim : 31.003 30.791 30.992 -> 31.003 -> 2015.171 MByte/s p14 random-cyc-1dim : 37.084 36.665 37.017 -> 37.084 -> 2410.487 MByte/s p15 random-cyc-1dim : 35.921 35.336 35.636 -> 35.921 -> 2334.848 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 21.916 22.174 21.945 -> 22.174 -> 1441.338 MByte/s p17 best bi-section : 68.073 67.852 68.600 -> 68.600 -> 4458.996 MByte/s p18 worst bi-section : 20.193 20.165 20.188 -> 20.193 -> 1312.513 MByte/s p19 acyclic-1dim-all : 64.926 64.590 64.349 -> 64.926 -> 4220.160 MByte/s p20 acyclic-2dim-all : 47.303 47.283 47.253 -> 47.303 -> 3074.705 MByte/s p21 acyclic-3dim-all : 43.096 42.617 42.875 -> 43.096 -> 2801.236 MByte/s p22 cyclic-1dim-all : 75.128 75.463 75.548 -> 75.548 -> 4910.643 MByte/s p23 cyclic-2dim-all : 47.135 47.125 47.903 -> 47.903 -> 3113.726 MByte/s p24 cyclic-3dim-all : 43.208 43.594 43.603 -> 43.603 -> 2834.214 MByte/s log_avg of all rings : 70.762 70.297 70.977 || 71.042 -> 4617.708 MByte/s log_avg of all random : 34.681 34.403 34.457 || 34.734 -> 2257.702 MByte/s log_avg(ring,random) : 49.539 49.178 49.453 ||( 49.674 -> 3228.840)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-32*2&+1 p00 method 0 : 0.052 0.824 9.598 50.808 171.208 165.036 -> 68.781 -> 4470.775 MByte/s p00 method 1 : 0.016 0.278 4.284 53.309 178.608 207.347 -> 69.624 -> 4525.565 MByte/s p00 method 2 : 0.023 0.388 4.693 26.088 146.564 122.606 -> 52.987 -> 3444.161 MByte/s p01 ring-16*4&+1 p01 method 0 : 0.051 0.817 9.327 50.072 142.473 153.740 -> 56.644 -> 3681.857 MByte/s p01 method 1 : 0.027 0.469 7.026 75.831 184.017 195.922 -> 74.562 -> 4846.528 MByte/s p01 method 2 : 0.021 0.338 4.403 29.340 120.675 98.412 -> 43.986 -> 2859.120 MByte/s p02 ring-8*8&+1 p02 method 0 : 0.050 0.803 8.923 48.496 134.468 138.692 -> 53.388 -> 3470.197 MByte/s p02 method 1 : 0.028 0.472 7.072 73.380 153.153 186.192 -> 66.445 -> 4318.896 MByte/s p02 method 2 : 0.021 0.340 4.414 29.729 116.570 125.721 -> 45.327 -> 2946.260 MByte/s p03 ring-4*16&+1 p03 method 0 : 0.048 0.764 8.240 40.394 132.074 129.235 -> 51.048 -> 3318.100 MByte/s p03 method 1 : 0.027 0.468 7.024 73.246 142.458 175.228 -> 64.628 -> 4200.796 MByte/s p03 method 2 : 0.021 0.344 4.463 29.619 113.235 111.245 -> 43.203 -> 2808.219 MByte/s p04 ring-2*32&+1 p04 method 0 : 0.048 0.780 8.306 40.165 142.137 132.274 -> 54.473 -> 3540.738 MByte/s p04 method 1 : 0.027 0.472 7.067 75.790 189.761 211.234 -> 77.080 -> 5010.222 MByte/s p04 method 2 : 0.022 0.344 4.513 29.811 125.392 112.743 -> 46.390 -> 3015.380 MByte/s p05 ring-1*65fix p05 method 0 : 0.047 0.769 8.344 38.717 144.408 130.026 -> 54.620 -> 3550.274 MByte/s p05 method 1 : 0.027 0.469 7.016 73.229 190.217 212.128 -> 76.303 -> 4959.671 MByte/s p05 method 2 : 0.022 0.343 4.464 29.982 125.610 110.362 -> 46.458 -> 3019.772 MByte/s p06 random-cyc-1dim p06 method 0 : 0.044 0.696 7.814 36.031 76.398 61.630 -> 32.645 -> 2121.893 MByte/s p06 method 1 : 0.026 0.457 6.865 64.206 73.020 45.345 -> 33.424 -> 2172.590 MByte/s p06 method 2 : 0.021 0.332 4.271 29.028 75.779 55.957 -> 28.864 -> 1876.168 MByte/s p07 random-cyc-1dim p07 method 0 : 0.043 0.696 7.788 37.743 82.864 62.816 -> 34.033 -> 2212.171 MByte/s p07 method 1 : 0.027 0.467 7.017 65.631 84.013 81.045 -> 39.121 -> 2542.864 MByte/s p07 method 2 : 0.021 0.330 4.297 29.115 72.326 66.059 -> 30.367 -> 1973.838 MByte/s p08 random-cyc-1dim p08 method 0 : 0.044 0.698 7.657 36.675 73.480 55.181 -> 30.896 -> 2008.241 MByte/s p08 method 1 : 0.027 0.469 7.028 59.259 70.367 47.355 -> 32.380 -> 2104.730 MByte/s p08 method 2 : 0.021 0.334 4.295 28.950 67.403 61.052 -> 28.075 -> 1824.893 MByte/s p09 random-cyc-1dim p09 method 0 : 0.044 0.693 7.814 36.084 76.868 61.244 -> 31.425 -> 2042.624 MByte/s p09 method 1 : 0.027 0.461 6.971 62.806 69.724 42.539 -> 31.952 -> 2076.905 MByte/s p09 method 2 : 0.021 0.335 4.341 29.086 68.612 54.983 -> 28.076 -> 1824.949 MByte/s p10 random-cyc-1dim p10 method 0 : 0.044 0.709 7.749 36.377 62.337 50.609 -> 27.751 -> 1803.811 MByte/s p10 method 1 : 0.027 0.466 6.978 56.357 61.276 38.377 -> 28.005 -> 1820.322 MByte/s p10 method 2 : 0.021 0.331 4.301 29.066 58.251 58.863 -> 25.558 -> 1661.255 MByte/s p11 random-cyc-1dim p11 method 0 : 0.044 0.698 7.726 36.508 78.834 67.870 -> 33.219 -> 2159.261 MByte/s p11 method 1 : 0.027 0.459 6.947 64.186 73.826 41.404 -> 32.600 -> 2119.029 MByte/s p11 method 2 : 0.021 0.333 4.306 29.021 71.189 59.560 -> 28.864 -> 1876.135 MByte/s p12 random-cyc-1dim p12 method 0 : 0.044 0.703 7.744 36.372 80.632 57.553 -> 32.720 -> 2126.831 MByte/s p12 method 1 : 0.027 0.465 7.003 65.634 72.715 54.310 -> 33.538 -> 2179.950 MByte/s p12 method 2 : 0.021 0.334 4.294 29.094 72.700 56.611 -> 29.402 -> 1911.153 MByte/s p13 random-cyc-1dim p13 method 0 : 0.044 0.710 7.691 35.976 66.130 58.866 -> 28.916 -> 1879.549 MByte/s p13 method 1 : 0.027 0.464 6.985 56.747 61.699 44.194 -> 29.205 -> 1898.356 MByte/s p13 method 2 : 0.021 0.334 4.277 28.901 61.424 60.124 -> 25.970 -> 1688.056 MByte/s p14 random-cyc-1dim p14 method 0 : 0.043 0.702 7.751 37.131 81.942 69.691 -> 34.028 -> 2211.843 MByte/s p14 method 1 : 0.028 0.467 6.990 61.884 77.280 65.001 -> 34.843 -> 2264.781 MByte/s p14 method 2 : 0.021 0.334 4.335 29.135 73.677 72.583 -> 30.763 -> 1999.597 MByte/s p15 random-cyc-1dim p15 method 0 : 0.044 0.699 7.858 36.999 78.690 64.284 -> 32.914 -> 2139.382 MByte/s p15 method 1 : 0.027 0.464 6.980 62.950 75.919 34.996 -> 32.354 -> 2103.004 MByte/s p15 method 2 : 0.021 0.337 4.322 28.622 69.165 62.276 -> 28.510 -> 1853.155 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.043 0.689 7.573 32.462 43.185 37.084 -> 20.898 -> 1358.387 MByte/s p16 method 1 : 0.027 0.466 7.033 40.944 42.849 35.892 -> 21.619 -> 1405.208 MByte/s p16 method 2 : 0.021 0.333 4.172 27.554 42.652 38.080 -> 19.585 -> 1273.002 MByte/s p17 best bi-section p17 method 0 : 0.032 0.491 5.450 27.585 129.613 160.275 -> 50.768 -> 3299.928 MByte/s p17 method 1 : 0.016 0.270 4.154 51.659 174.873 205.217 -> 68.602 -> 4459.119 MByte/s p17 method 2 : 0.015 0.233 3.323 23.982 123.977 159.106 -> 47.984 -> 3118.932 MByte/s p18 worst bi-section p18 method 0 : 0.023 0.362 4.442 24.790 38.696 39.404 -> 17.809 -> 1157.613 MByte/s p18 method 1 : 0.016 0.265 4.059 32.763 38.123 55.510 -> 20.210 -> 1313.667 MByte/s p18 method 2 : 0.015 0.237 3.157 21.914 36.360 34.396 -> 16.497 -> 1072.294 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.047 0.758 8.327 37.318 141.073 130.591 -> 54.219 -> 3524.256 MByte/s p19 method 1 : 0.027 0.467 6.973 74.940 165.611 144.398 -> 65.373 -> 4249.275 MByte/s p19 method 2 : 0.021 0.326 4.289 29.685 124.644 109.327 -> 45.740 -> 2973.109 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.035 0.552 6.049 28.054 105.110 103.437 -> 39.780 -> 2585.694 MByte/s p20 method 1 : 0.038 0.663 9.509 75.939 100.575 85.749 -> 46.021 -> 2991.346 MByte/s p20 method 2 : 0.018 0.294 3.775 25.650 89.859 78.525 -> 34.306 -> 2229.907 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.028 0.472 5.213 23.938 91.623 94.334 -> 35.862 -> 2331.060 MByte/s p21 method 1 : 0.042 0.752 10.581 70.764 81.617 72.134 -> 38.683 -> 2514.377 MByte/s p21 method 2 : 0.016 0.252 3.253 21.913 82.529 64.789 -> 31.089 -> 2020.773 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.047 0.767 8.321 38.681 141.127 132.275 -> 54.357 -> 3533.229 MByte/s p22 method 1 : 0.027 0.464 6.912 74.147 189.514 211.709 -> 76.030 -> 4941.974 MByte/s p22 method 2 : 0.022 0.336 4.454 29.824 124.447 102.799 -> 45.658 -> 2967.795 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.039 0.615 6.750 31.782 104.083 103.214 -> 40.978 -> 2663.585 MByte/s p23 method 1 : 0.041 0.714 10.392 75.192 101.195 80.596 -> 45.849 -> 2980.175 MByte/s p23 method 2 : 0.021 0.339 4.343 29.374 92.369 76.937 -> 35.695 -> 2320.207 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.037 0.589 6.411 30.518 90.326 90.416 -> 36.570 -> 2377.028 MByte/s p24 method 1 : 0.052 0.924 13.023 70.998 83.788 75.885 -> 41.075 -> 2669.849 MByte/s p24 method 2 : 0.021 0.333 4.169 28.231 78.455 67.503 -> 31.156 -> 2025.142 MByte/s log_avg of all rings - ring, method 0 : 0.049 0.792 8.774 44.485 143.935 140.891 || 56.225 -> 3654.614 MByte/s - ring, method 1 : 0.025 0.431 6.481 70.289 171.991 197.523 || 71.275 -> 4632.900 MByte/s - ring, method 2 : 0.022 0.349 4.491 29.061 124.237 113.160 || 46.289 -> 3008.770 MByte/s log_avg of all random - random, method 0 : 0.044 0.700 7.759 36.586 75.529 60.727 || 31.789 -> 2066.271 MByte/s - random, method 1 : 0.027 0.464 6.976 61.880 71.686 47.948 || 32.619 -> 2120.259 MByte/s - random, method 2 : 0.021 0.333 4.304 29.001 68.845 60.612 || 28.400 -> 1845.996 MByte/s log_avg(ring,random) - average, method 0 : 0.046 0.745 8.251 40.343 104.265 92.498 || 42.277 -> 2747.985 MByte/s - average, method 1 : 0.026 0.447 6.724 65.950 111.038 97.318 || 48.218 -> 3134.158 MByte/s - average, method 2 : 0.021 0.341 4.396 29.031 92.483 82.818 || 36.257 -> 2356.731 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 3.021 0.046 0.049 0.044 0.046 0.026 0.021 2 6.083 0.094 0.099 0.089 0.094 0.052 0.043 4 12.127 0.187 0.197 0.176 0.187 0.114 0.086 8 24.265 0.373 0.396 0.352 0.373 0.226 0.172 16 48.416 0.745 0.792 0.700 0.745 0.447 0.341 32 73.881 1.137 1.246 1.037 1.137 0.878 0.591 64 135.126 2.079 2.253 1.918 2.079 1.697 1.130 128 294.413 4.529 4.904 4.184 4.529 3.498 2.332 256 536.298 8.251 8.774 7.759 8.251 6.724 4.396 512 961.833 14.797 15.688 13.958 14.797 12.889 8.038 1024 1666.101 25.632 26.257 25.023 22.725 24.328 13.768 2048 2813.259 43.281 43.738 42.829 31.317 42.517 21.355 4096 4286.762 65.950 70.289 61.880 40.343 65.950 29.031 8192 5596.139 86.094 104.129 71.183 62.275 86.062 59.701 16384 6471.716 99.565 135.482 73.169 81.955 99.506 76.768 32768 7035.172 108.233 159.223 73.573 96.301 107.444 87.266 65536 7413.481 114.054 171.991 75.633 104.265 111.038 92.483 131072 7575.701 116.549 180.335 75.325 106.999 111.420 92.658 262144 7724.499 118.838 188.890 74.766 106.617 109.497 93.495 524288 7606.553 117.024 194.373 70.455 103.673 105.082 90.480 1048576 7324.285 112.681 197.523 64.281 92.498 97.318 82.818 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-32*2&+1 : 0.052 0.824 9.598 53.309 178.608 207.347 -> 71.620 -> 4655.307 MByte/s p01 ring-16*4&+1 : 0.051 0.817 9.327 75.831 184.017 195.922 -> 75.079 -> 4880.132 MByte/s p02 ring-8*8&+1 : 0.050 0.803 8.923 73.380 153.153 186.192 -> 66.850 -> 4345.251 MByte/s p03 ring-4*16&+1 : 0.048 0.764 8.240 73.246 142.458 175.228 -> 64.851 -> 4215.346 MByte/s p04 ring-2*32&+1 : 0.048 0.780 8.306 75.790 189.761 211.234 -> 77.329 -> 5026.353 MByte/s p05 ring-1*65fix : 0.047 0.769 8.344 73.229 190.217 212.128 -> 76.517 -> 4973.612 MByte/s p06 random-cyc-1dim : 0.044 0.696 7.814 64.206 76.398 61.630 -> 35.965 -> 2337.738 MByte/s p07 random-cyc-1dim : 0.043 0.696 7.788 65.631 84.013 81.045 -> 39.247 -> 2551.074 MByte/s p08 random-cyc-1dim : 0.044 0.698 7.657 59.259 73.480 61.052 -> 34.096 -> 2216.210 MByte/s p09 random-cyc-1dim : 0.044 0.693 7.814 62.806 76.868 61.244 -> 34.692 -> 2254.979 MByte/s p10 random-cyc-1dim : 0.044 0.709 7.749 56.357 62.337 58.863 -> 30.528 -> 1984.331 MByte/s p11 random-cyc-1dim : 0.044 0.698 7.726 64.186 78.834 67.870 -> 36.824 -> 2393.561 MByte/s p12 random-cyc-1dim : 0.044 0.703 7.744 65.634 80.632 57.553 -> 36.284 -> 2358.470 MByte/s p13 random-cyc-1dim : 0.044 0.710 7.691 56.747 66.130 60.124 -> 31.439 -> 2043.561 MByte/s p14 random-cyc-1dim : 0.043 0.702 7.751 61.884 81.942 72.583 -> 37.491 -> 2436.885 MByte/s p15 random-cyc-1dim : 0.044 0.699 7.858 62.950 78.690 64.284 -> 36.290 -> 2358.854 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.043 0.689 7.573 40.944 43.185 38.080 -> 22.333 -> 1451.629 MByte/s p17 best bi-section : 0.032 0.491 5.450 51.659 174.873 205.217 -> 68.824 -> 4473.563 MByte/s p18 worst bi-section : 0.023 0.362 4.442 32.763 38.696 55.510 -> 20.308 -> 1320.045 MByte/s p19 acyclic-1dim-all : 0.047 0.758 8.327 74.940 165.611 144.398 -> 65.693 -> 4270.067 MByte/s p20 acyclic-2dim-all : 0.038 0.663 9.509 75.939 105.110 103.437 -> 48.008 -> 3120.494 MByte/s p21 acyclic-3dim-all : 0.042 0.752 10.581 70.764 91.623 94.334 -> 43.507 -> 2827.957 MByte/s p22 cyclic-1dim-all : 0.047 0.767 8.321 74.147 189.514 211.709 -> 76.210 -> 4953.619 MByte/s p23 cyclic-2dim-all : 0.041 0.714 10.392 75.192 104.083 103.214 -> 48.411 -> 3146.740 MByte/s p24 cyclic-3dim-all : 0.052 0.924 13.023 70.998 90.326 90.416 -> 43.775 -> 2845.380 MByte/s log_avg of all rings : 0.049 0.792 8.774 70.289 171.991 197.523 || 71.881 -> 4672.244 MByte/s log_avg of all random : 0.044 0.700 7.759 61.880 75.633 64.281 || 35.191 -> 2287.434 MByte/s log_avg(ring,random) : 0.046 0.745 8.251 65.950 114.054 112.681 || 50.295 -> 3269.166 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 3269.166 MByte/s on 65 processes ( = 50.295 MByte/s * 65 processes) system parameters : 65 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 = 3269.166 MB/s = 50.295 * 65 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E