b_eff = 182.989 MB/s = 91.495 * 2 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 = 2 1-dim-paterns: size = 2 2-dim-paterns: size = 2 * 1 3-dim-paterns: size = 2 * 1 * 1 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 = 46.174 sec sum of max elapsed time per entries above = 46.046 sec difference = 0.128 sec = 0.3% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-1*2fix => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p01 ring-1*2fix => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p02 ring-1*2fix => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p03 ring-1*2fix => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p04 ring-1*2fix => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p05 ring-1*2fix => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 2 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 2 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p22 cyclic-1dim-all => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 1 sendrecv_calls with 2 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 1 sendrecv_calls with 2 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-1*2fix : 72.718 91.530 65.852 -> 91.530 -> 183.060 MByte/s p01 ring-1*2fix : 72.874 91.614 65.720 -> 91.614 -> 183.228 MByte/s p02 ring-1*2fix : 72.621 91.464 65.896 -> 91.464 -> 182.927 MByte/s p03 ring-1*2fix : 72.642 91.337 65.932 -> 91.337 -> 182.675 MByte/s p04 ring-1*2fix : 72.670 91.494 65.841 -> 91.494 -> 182.988 MByte/s p05 ring-1*2fix : 72.665 91.432 65.998 -> 91.432 -> 182.863 MByte/s p06 random-cyc-1dim : 72.623 91.631 65.873 -> 91.631 -> 183.263 MByte/s p07 random-cyc-1dim : 72.694 91.485 65.843 -> 91.485 -> 182.970 MByte/s p08 random-cyc-1dim : 72.659 91.347 65.831 -> 91.347 -> 182.693 MByte/s p09 random-cyc-1dim : 72.633 91.473 65.868 -> 91.473 -> 182.945 MByte/s p10 random-cyc-1dim : 72.615 91.642 65.914 -> 91.642 -> 183.284 MByte/s p11 random-cyc-1dim : 72.743 91.499 65.750 -> 91.499 -> 182.998 MByte/s p12 random-cyc-1dim : 72.626 91.624 65.779 -> 91.624 -> 183.248 MByte/s p13 random-cyc-1dim : 72.739 91.527 65.879 -> 91.527 -> 183.053 MByte/s p14 random-cyc-1dim : 72.770 91.399 65.780 -> 91.399 -> 182.797 MByte/s p15 random-cyc-1dim : 72.811 91.480 65.773 -> 91.480 -> 182.961 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 72.610 91.368 65.718 -> 91.368 -> 182.735 MByte/s p17 best bi-section : 52.234 91.237 51.993 -> 91.237 -> 182.474 MByte/s p18 worst bi-section : 52.033 91.193 51.727 -> 91.193 -> 182.387 MByte/s p19 acyclic-1dim-all : 52.242 91.246 51.613 -> 91.246 -> 182.492 MByte/s p20 acyclic-2dim-all : 52.182 91.260 51.203 -> 91.260 -> 182.519 MByte/s p21 acyclic-3dim-all : 52.205 91.266 50.702 -> 91.266 -> 182.532 MByte/s p22 cyclic-1dim-all : 72.790 90.739 66.007 -> 90.739 -> 181.477 MByte/s p23 cyclic-2dim-all : 72.752 90.771 66.074 -> 90.771 -> 181.541 MByte/s p24 cyclic-3dim-all : 72.734 90.587 65.865 -> 90.587 -> 181.175 MByte/s log_avg of all rings : 72.698 91.478 65.873 || 91.478 -> 182.957 MByte/s log_avg of all random : 72.691 91.511 65.829 || 91.511 -> 183.021 MByte/s log_avg(ring,random) : 72.695 91.495 65.851 ||( 91.495 -> 182.989)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-1*2fix : 91.114 91.029 91.272 -> 91.272 -> 182.544 MByte/s p01 ring-1*2fix : 91.056 91.237 91.141 -> 91.237 -> 182.474 MByte/s p02 ring-1*2fix : 91.209 91.054 91.214 -> 91.214 -> 182.428 MByte/s p03 ring-1*2fix : 90.486 91.176 91.029 -> 91.176 -> 182.351 MByte/s p04 ring-1*2fix : 91.222 91.183 91.254 -> 91.254 -> 182.508 MByte/s p05 ring-1*2fix : 91.130 91.160 91.085 -> 91.160 -> 182.320 MByte/s p06 random-cyc-1dim : 91.195 91.256 91.293 -> 91.293 -> 182.587 MByte/s p07 random-cyc-1dim : 91.066 89.349 91.328 -> 91.328 -> 182.655 MByte/s p08 random-cyc-1dim : 91.151 91.158 91.057 -> 91.158 -> 182.316 MByte/s p09 random-cyc-1dim : 91.159 90.527 91.224 -> 91.224 -> 182.448 MByte/s p10 random-cyc-1dim : 91.440 91.197 91.099 -> 91.440 -> 182.881 MByte/s p11 random-cyc-1dim : 91.110 91.298 91.210 -> 91.298 -> 182.595 MByte/s p12 random-cyc-1dim : 91.256 91.053 91.294 -> 91.294 -> 182.588 MByte/s p13 random-cyc-1dim : 90.926 91.160 91.187 -> 91.187 -> 182.374 MByte/s p14 random-cyc-1dim : 90.566 91.231 90.464 -> 91.231 -> 182.463 MByte/s p15 random-cyc-1dim : 91.311 91.269 91.216 -> 91.311 -> 182.621 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 91.214 91.097 91.163 -> 91.214 -> 182.428 MByte/s p17 best bi-section : 90.731 91.041 90.886 -> 91.041 -> 182.083 MByte/s p18 worst bi-section : 90.827 90.874 90.791 -> 90.874 -> 181.748 MByte/s p19 acyclic-1dim-all : 90.869 90.942 90.832 -> 90.942 -> 181.884 MByte/s p20 acyclic-2dim-all : 90.901 91.021 91.110 -> 91.110 -> 182.220 MByte/s p21 acyclic-3dim-all : 90.694 91.001 90.953 -> 91.001 -> 182.002 MByte/s p22 cyclic-1dim-all : 90.354 90.521 90.491 -> 90.521 -> 181.042 MByte/s p23 cyclic-2dim-all : 90.503 90.337 90.491 -> 90.503 -> 181.007 MByte/s p24 cyclic-3dim-all : 90.353 90.319 90.296 -> 90.353 -> 180.706 MByte/s log_avg of all rings : 91.036 91.140 91.166 || 91.219 -> 182.438 MByte/s log_avg of all random : 91.118 90.948 91.137 || 91.276 -> 182.553 MByte/s log_avg(ring,random) : 91.077 91.044 91.151 ||( 91.248 -> 182.495)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-1*2fix p00 method 0 : 0.057 0.882 10.532 52.254 180.733 208.206 -> 72.718 -> 145.437 MByte/s p00 method 1 : 0.065 1.370 18.517 128.545 203.096 210.111 -> 91.530 -> 183.060 MByte/s p00 method 2 : 0.026 0.407 5.454 27.255 169.433 207.803 -> 65.852 -> 131.703 MByte/s p01 ring-1*2fix p01 method 0 : 0.057 0.883 10.383 52.200 180.507 209.481 -> 72.874 -> 145.748 MByte/s p01 method 1 : 0.066 1.364 18.503 127.863 202.624 211.089 -> 91.614 -> 183.228 MByte/s p01 method 2 : 0.026 0.410 5.454 27.092 169.511 207.323 -> 65.720 -> 131.441 MByte/s p02 ring-1*2fix p02 method 0 : 0.056 0.864 10.377 52.073 180.001 208.137 -> 72.621 -> 145.242 MByte/s p02 method 1 : 0.066 1.374 18.557 126.999 201.113 210.438 -> 91.464 -> 182.927 MByte/s p02 method 2 : 0.026 0.402 5.416 27.019 170.272 209.750 -> 65.896 -> 131.791 MByte/s p03 ring-1*2fix p03 method 0 : 0.056 0.878 10.555 52.148 179.684 207.744 -> 72.642 -> 145.284 MByte/s p03 method 1 : 0.065 1.369 18.109 128.361 202.321 209.546 -> 91.337 -> 182.675 MByte/s p03 method 2 : 0.026 0.409 5.454 27.425 169.586 208.196 -> 65.932 -> 131.865 MByte/s p04 ring-1*2fix p04 method 0 : 0.057 0.888 10.370 52.308 179.637 208.453 -> 72.670 -> 145.340 MByte/s p04 method 1 : 0.065 1.369 18.761 126.966 202.121 210.644 -> 91.494 -> 182.988 MByte/s p04 method 2 : 0.026 0.409 5.424 27.416 169.835 206.957 -> 65.841 -> 131.682 MByte/s p05 ring-1*2fix p05 method 0 : 0.056 0.886 10.524 52.360 180.366 208.473 -> 72.665 -> 145.330 MByte/s p05 method 1 : 0.066 1.338 18.497 126.553 202.956 210.156 -> 91.432 -> 182.863 MByte/s p05 method 2 : 0.026 0.408 5.461 27.600 169.707 208.049 -> 65.998 -> 131.996 MByte/s p06 random-cyc-1dim p06 method 0 : 0.056 0.884 10.426 52.230 180.300 208.315 -> 72.623 -> 145.246 MByte/s p06 method 1 : 0.065 1.370 18.694 128.152 202.694 209.881 -> 91.631 -> 183.263 MByte/s p06 method 2 : 0.026 0.411 5.475 28.704 169.299 207.382 -> 65.873 -> 131.746 MByte/s p07 random-cyc-1dim p07 method 0 : 0.056 0.887 10.427 52.125 180.807 209.157 -> 72.694 -> 145.389 MByte/s p07 method 1 : 0.065 1.342 18.453 128.068 202.158 210.357 -> 91.485 -> 182.970 MByte/s p07 method 2 : 0.026 0.409 5.458 27.158 169.819 207.172 -> 65.843 -> 131.686 MByte/s p08 random-cyc-1dim p08 method 0 : 0.056 0.877 10.438 52.014 180.237 208.572 -> 72.659 -> 145.318 MByte/s p08 method 1 : 0.064 1.354 18.732 126.583 201.450 210.237 -> 91.347 -> 182.693 MByte/s p08 method 2 : 0.026 0.404 5.491 27.552 169.937 207.631 -> 65.831 -> 131.661 MByte/s p09 random-cyc-1dim p09 method 0 : 0.057 0.884 10.425 52.172 179.736 208.251 -> 72.633 -> 145.266 MByte/s p09 method 1 : 0.069 1.359 18.531 126.450 202.647 210.831 -> 91.473 -> 182.945 MByte/s p09 method 2 : 0.026 0.404 5.486 26.795 170.421 208.532 -> 65.868 -> 131.735 MByte/s p10 random-cyc-1dim p10 method 0 : 0.056 0.883 10.501 52.132 180.126 208.285 -> 72.615 -> 145.230 MByte/s p10 method 1 : 0.069 1.361 18.655 128.126 203.213 210.902 -> 91.642 -> 183.284 MByte/s p10 method 2 : 0.026 0.410 5.487 27.025 169.671 207.543 -> 65.914 -> 131.829 MByte/s p11 random-cyc-1dim p11 method 0 : 0.057 0.878 10.572 52.092 179.776 209.222 -> 72.743 -> 145.485 MByte/s p11 method 1 : 0.070 1.378 18.428 127.073 202.610 210.066 -> 91.499 -> 182.998 MByte/s p11 method 2 : 0.026 0.403 5.477 26.491 169.887 208.152 -> 65.750 -> 131.500 MByte/s p12 random-cyc-1dim p12 method 0 : 0.056 0.875 10.294 52.219 180.152 209.750 -> 72.626 -> 145.252 MByte/s p12 method 1 : 0.069 1.366 18.555 128.176 202.689 211.018 -> 91.624 -> 183.248 MByte/s p12 method 2 : 0.026 0.405 5.448 26.999 169.117 208.631 -> 65.779 -> 131.558 MByte/s p13 random-cyc-1dim p13 method 0 : 0.057 0.887 10.581 52.208 180.377 208.894 -> 72.739 -> 145.478 MByte/s p13 method 1 : 0.069 1.371 18.745 128.172 202.330 211.434 -> 91.527 -> 183.053 MByte/s p13 method 2 : 0.026 0.410 5.496 27.136 169.979 208.849 -> 65.879 -> 131.757 MByte/s p14 random-cyc-1dim p14 method 0 : 0.057 0.878 10.498 52.058 180.459 208.775 -> 72.770 -> 145.541 MByte/s p14 method 1 : 0.070 1.341 18.806 126.992 201.385 210.332 -> 91.399 -> 182.797 MByte/s p14 method 2 : 0.025 0.405 5.478 26.615 169.789 208.211 -> 65.780 -> 131.559 MByte/s p15 random-cyc-1dim p15 method 0 : 0.056 0.884 10.594 52.027 179.714 209.476 -> 72.811 -> 145.623 MByte/s p15 method 1 : 0.069 1.371 18.720 127.854 201.473 210.584 -> 91.480 -> 182.961 MByte/s p15 method 2 : 0.026 0.405 5.495 27.277 169.887 206.714 -> 65.773 -> 131.546 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.057 0.871 10.591 52.218 180.100 208.587 -> 72.610 -> 145.219 MByte/s p16 method 1 : 0.069 1.370 18.500 127.100 201.958 210.463 -> 91.368 -> 182.735 MByte/s p16 method 2 : 0.026 0.406 5.374 27.698 169.211 208.152 -> 65.718 -> 131.436 MByte/s p17 best bi-section p17 method 0 : 0.034 0.529 5.844 28.622 131.860 163.880 -> 52.234 -> 104.468 MByte/s p17 method 1 : 0.068 1.331 17.712 124.362 201.921 211.404 -> 91.237 -> 182.474 MByte/s p17 method 2 : 0.015 0.243 3.610 28.268 129.233 162.463 -> 51.993 -> 103.986 MByte/s p18 worst bi-section p18 method 0 : 0.035 0.529 5.813 28.604 132.137 163.691 -> 52.033 -> 104.066 MByte/s p18 method 1 : 0.068 1.329 17.650 124.095 202.521 211.105 -> 91.193 -> 182.387 MByte/s p18 method 2 : 0.015 0.246 3.641 28.220 130.461 163.205 -> 51.727 -> 103.453 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.035 0.515 5.850 28.608 133.389 163.190 -> 52.242 -> 104.484 MByte/s p19 method 1 : 0.067 1.325 17.768 124.714 201.982 210.352 -> 91.246 -> 182.492 MByte/s p19 method 2 : 0.016 0.244 3.633 28.078 128.468 162.367 -> 51.613 -> 103.226 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.035 0.515 5.804 28.827 132.089 163.548 -> 52.182 -> 104.363 MByte/s p20 method 1 : 0.069 1.329 17.692 124.413 201.833 211.333 -> 91.260 -> 182.519 MByte/s p20 method 2 : 0.015 0.252 3.589 28.242 131.763 162.418 -> 51.203 -> 102.405 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.034 0.525 5.815 28.576 133.262 163.175 -> 52.205 -> 104.409 MByte/s p21 method 1 : 0.068 1.317 17.774 125.502 201.833 210.031 -> 91.266 -> 182.532 MByte/s p21 method 2 : 0.016 0.248 3.618 28.039 131.844 162.541 -> 50.702 -> 101.403 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.057 0.882 10.602 52.296 180.322 208.502 -> 72.790 -> 145.580 MByte/s p22 method 1 : 0.065 1.293 17.541 124.441 201.657 210.287 -> 90.739 -> 181.477 MByte/s p22 method 2 : 0.026 0.403 5.510 27.059 169.914 209.426 -> 66.007 -> 132.015 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.057 0.887 10.555 52.117 180.089 208.894 -> 72.752 -> 145.503 MByte/s p23 method 1 : 0.065 1.300 17.693 125.535 202.256 210.352 -> 90.771 -> 181.541 MByte/s p23 method 2 : 0.026 0.409 5.484 28.077 170.108 208.300 -> 66.074 -> 132.148 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.057 0.887 10.600 52.189 179.552 209.830 -> 72.734 -> 145.467 MByte/s p24 method 1 : 0.066 1.269 17.569 124.246 201.792 210.589 -> 90.587 -> 181.175 MByte/s p24 method 2 : 0.026 0.405 5.493 28.737 169.681 208.039 -> 65.865 -> 131.731 MByte/s log_avg of all rings - ring, method 0 : 0.056 0.880 10.456 52.224 180.154 208.415 || 72.698 -> 145.397 MByte/s - ring, method 1 : 0.066 1.364 18.490 127.546 202.371 210.330 || 91.478 -> 182.957 MByte/s - ring, method 2 : 0.026 0.407 5.444 27.300 169.724 208.011 || 65.873 -> 131.746 MByte/s log_avg of all random - random, method 0 : 0.056 0.882 10.475 52.128 180.168 208.869 || 72.691 -> 145.383 MByte/s - random, method 1 : 0.068 1.361 18.632 127.563 202.264 210.564 || 91.511 -> 183.021 MByte/s - random, method 2 : 0.026 0.407 5.479 27.169 169.780 207.881 || 65.829 -> 131.658 MByte/s log_avg(ring,random) - average, method 0 : 0.056 0.881 10.466 52.176 180.161 208.642 || 72.695 -> 145.390 MByte/s - average, method 1 : 0.067 1.363 18.560 127.554 202.317 210.447 || 91.495 -> 182.989 MByte/s - average, method 2 : 0.026 0.407 5.461 27.235 169.752 207.946 || 65.851 -> 131.702 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 0.133 0.067 0.066 0.068 0.056 0.067 0.026 2 0.266 0.133 0.130 0.136 0.113 0.133 0.051 4 0.712 0.356 0.357 0.356 0.226 0.356 0.102 8 1.398 0.699 0.698 0.700 0.450 0.699 0.205 16 2.725 1.363 1.364 1.361 0.881 1.363 0.407 32 5.277 2.638 2.634 2.643 1.461 2.638 0.738 64 9.561 4.781 4.778 4.783 2.665 4.781 1.391 128 20.491 10.245 10.255 10.236 5.736 10.245 2.869 256 37.121 18.560 18.490 18.632 10.466 18.560 5.461 512 66.363 33.181 33.079 33.284 18.349 33.181 9.908 1024 115.022 57.511 57.507 57.514 29.542 57.511 17.562 2048 182.680 91.340 91.401 91.280 40.323 91.340 26.532 4096 255.109 127.554 127.546 127.563 52.176 127.554 27.235 8192 318.639 159.320 159.387 159.252 89.469 159.320 72.089 16384 362.143 181.071 181.232 180.911 125.502 181.071 107.267 32768 389.381 194.690 194.783 194.598 157.562 194.690 141.924 65536 404.634 202.317 202.371 202.264 180.161 202.317 169.752 131072 412.789 206.395 206.374 206.415 194.449 206.395 188.047 262144 417.822 208.911 208.497 209.325 201.980 208.911 198.761 524288 419.586 209.793 209.758 209.828 206.378 209.793 204.587 1048576 420.894 210.447 210.330 210.564 208.642 210.447 207.946 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-1*2fix : 0.065 1.370 18.517 128.545 203.096 210.111 -> 91.530 -> 183.060 MByte/s p01 ring-1*2fix : 0.066 1.364 18.503 127.863 202.624 211.089 -> 91.614 -> 183.228 MByte/s p02 ring-1*2fix : 0.066 1.374 18.557 126.999 201.113 210.438 -> 91.464 -> 182.927 MByte/s p03 ring-1*2fix : 0.065 1.369 18.109 128.361 202.321 209.546 -> 91.337 -> 182.675 MByte/s p04 ring-1*2fix : 0.065 1.369 18.761 126.966 202.121 210.644 -> 91.494 -> 182.988 MByte/s p05 ring-1*2fix : 0.066 1.338 18.497 126.553 202.956 210.156 -> 91.432 -> 182.863 MByte/s p06 random-cyc-1dim : 0.065 1.370 18.694 128.152 202.694 209.881 -> 91.631 -> 183.263 MByte/s p07 random-cyc-1dim : 0.065 1.342 18.453 128.068 202.158 210.357 -> 91.485 -> 182.970 MByte/s p08 random-cyc-1dim : 0.064 1.354 18.732 126.583 201.450 210.237 -> 91.347 -> 182.693 MByte/s p09 random-cyc-1dim : 0.069 1.359 18.531 126.450 202.647 210.831 -> 91.473 -> 182.945 MByte/s p10 random-cyc-1dim : 0.069 1.361 18.655 128.126 203.213 210.902 -> 91.642 -> 183.284 MByte/s p11 random-cyc-1dim : 0.070 1.378 18.428 127.073 202.610 210.066 -> 91.499 -> 182.998 MByte/s p12 random-cyc-1dim : 0.069 1.366 18.555 128.176 202.689 211.018 -> 91.624 -> 183.248 MByte/s p13 random-cyc-1dim : 0.069 1.371 18.745 128.172 202.330 211.434 -> 91.527 -> 183.053 MByte/s p14 random-cyc-1dim : 0.070 1.341 18.806 126.992 201.385 210.332 -> 91.399 -> 182.797 MByte/s p15 random-cyc-1dim : 0.069 1.371 18.720 127.854 201.473 210.584 -> 91.480 -> 182.961 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.069 1.370 18.500 127.100 201.958 210.463 -> 91.368 -> 182.735 MByte/s p17 best bi-section : 0.068 1.331 17.712 124.362 201.921 211.404 -> 91.237 -> 182.474 MByte/s p18 worst bi-section : 0.068 1.329 17.650 124.095 202.521 211.105 -> 91.193 -> 182.387 MByte/s p19 acyclic-1dim-all : 0.067 1.325 17.768 124.714 201.982 210.352 -> 91.246 -> 182.492 MByte/s p20 acyclic-2dim-all : 0.069 1.329 17.692 124.413 201.833 211.333 -> 91.260 -> 182.519 MByte/s p21 acyclic-3dim-all : 0.068 1.317 17.774 125.502 201.833 210.031 -> 91.266 -> 182.532 MByte/s p22 cyclic-1dim-all : 0.065 1.293 17.541 124.441 201.657 210.287 -> 90.739 -> 181.477 MByte/s p23 cyclic-2dim-all : 0.065 1.300 17.693 125.535 202.256 210.352 -> 90.771 -> 181.541 MByte/s p24 cyclic-3dim-all : 0.066 1.269 17.569 124.246 201.792 210.589 -> 90.587 -> 181.175 MByte/s log_avg of all rings : 0.066 1.364 18.490 127.546 202.371 210.330 || 91.478 -> 182.957 MByte/s log_avg of all random : 0.068 1.361 18.632 127.563 202.264 210.564 || 91.511 -> 183.021 MByte/s log_avg(ring,random) : 0.067 1.363 18.560 127.554 202.317 210.447 || 91.495 -> 182.989 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 182.989 MByte/s on 2 processes ( = 91.495 MByte/s * 2 processes) system parameters : 2 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 = 182.989 MB/s = 91.495 * 2 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E