b_eff = 2999.941 MB/s = 53.570 * 56 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 = 56 1-dim-paterns: size = 56 2-dim-paterns: size = 8 * 7 3-dim-paterns: size = 7 * 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 = 149.039 sec sum of max elapsed time per entries above = 151.955 sec difference = -2.917 sec = 2.0% The difference is less than 5 % SECTION-LOOP-END SECTION-PATTERN-BEGIN Pattern parameters: p00 ring-28*2fix => 1 sendrecv_calls with 56 messages, i.e. msgs/used node, all nodes are used p01 ring-14*4fix => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p02 ring-7*8fix => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p03 ring-3*18&+1 => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p04 ring-2*28fix => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p05 ring-1*56fix => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p06 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p07 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p08 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p09 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p10 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p11 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p12 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p13 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p14 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p15 random-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p16 worst-cyc-1dim => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p17 best bi-section => 2 sendrecv_calls with 56 messages, i.e. msgs/used node, all nodes are used p18 worst bi-section => 2 sendrecv_calls with 56 messages, i.e. msgs/used node, all nodes are used p19 acyclic-1dim-all => 2 sendrecv_calls with 110 messages, i.e. msgs/used node, all nodes are used p20 acyclic-2dim-all => 4 sendrecv_calls with 194 messages, i.e. msgs/used node, all nodes are used p21 acyclic-3dim-all => 6 sendrecv_calls with 236 messages, i.e. msgs/used node, all nodes are used p22 cyclic-1dim-all => 2 sendrecv_calls with 112 messages, i.e. msgs/used node, all nodes are used p23 cyclic-2dim-all => 4 sendrecv_calls with 224 messages, i.e. msgs/used node, all nodes are used p24 cyclic-3dim-all => 5 sendrecv_calls with 280 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-28*2fix : 71.994 71.373 60.302 -> 71.994 -> 4031.667 MByte/s p01 ring-14*4fix : 59.408 75.798 45.293 -> 75.798 -> 4244.714 MByte/s p02 ring-7*8fix : 53.409 67.812 44.683 -> 67.812 -> 3797.461 MByte/s p03 ring-3*18&+1 : 52.811 64.998 44.702 -> 64.998 -> 3639.863 MByte/s p04 ring-2*28fix : 54.433 77.102 47.505 -> 77.102 -> 4317.714 MByte/s p05 ring-1*56fix : 53.610 71.161 44.699 -> 71.161 -> 3985.032 MByte/s p06 random-cyc-1dim : 29.077 30.700 26.841 -> 30.700 -> 1719.182 MByte/s p07 random-cyc-1dim : 33.794 36.941 31.068 -> 36.941 -> 2068.710 MByte/s p08 random-cyc-1dim : 33.980 35.855 30.519 -> 35.855 -> 2007.879 MByte/s p09 random-cyc-1dim : 35.792 40.129 32.037 -> 40.129 -> 2247.243 MByte/s p10 random-cyc-1dim : 41.534 44.902 35.157 -> 44.902 -> 2514.534 MByte/s p11 random-cyc-1dim : 38.443 40.610 33.414 -> 40.610 -> 2274.171 MByte/s p12 random-cyc-1dim : 31.275 33.842 29.052 -> 33.842 -> 1895.151 MByte/s p13 random-cyc-1dim : 33.294 33.660 29.725 -> 33.660 -> 1884.943 MByte/s p14 random-cyc-1dim : 38.750 42.654 34.807 -> 42.654 -> 2388.647 MByte/s p15 random-cyc-1dim : 40.799 47.135 36.351 -> 47.135 -> 2639.583 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 24.400 25.444 21.645 -> 25.444 -> 1424.839 MByte/s p17 best bi-section : 51.542 71.281 49.003 -> 71.281 -> 3991.759 MByte/s p18 worst bi-section : 18.185 20.980 17.040 -> 20.980 -> 1174.886 MByte/s p19 acyclic-1dim-all : 54.069 66.989 45.443 -> 66.989 -> 3751.368 MByte/s p20 acyclic-2dim-all : 43.146 51.589 36.521 -> 51.589 -> 2888.983 MByte/s p21 acyclic-3dim-all : 42.137 50.378 37.208 -> 50.378 -> 2821.188 MByte/s p22 cyclic-1dim-all : 54.031 71.158 44.811 -> 71.158 -> 3984.864 MByte/s p23 cyclic-2dim-all : 46.812 50.197 38.749 -> 50.197 -> 2811.039 MByte/s p24 cyclic-3dim-all : 46.583 53.472 41.583 -> 53.472 -> 2994.457 MByte/s log_avg of all rings : 57.249 71.250 47.568 || 71.353 -> 3995.761 MByte/s log_avg of all random : 35.458 38.313 31.767 || 38.313 -> 2145.517 MByte/s log_avg(ring,random) : 45.054 52.247 38.873 ||( 52.285 -> 2927.964)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-28*2fix : 72.653 72.797 72.700 -> 72.797 -> 4076.660 MByte/s p01 ring-14*4fix : 75.273 75.890 73.424 -> 75.890 -> 4249.844 MByte/s p02 ring-7*8fix : 67.098 67.209 66.798 -> 67.209 -> 3763.694 MByte/s p03 ring-3*18&+1 : 63.985 64.304 63.641 -> 64.304 -> 3601.012 MByte/s p04 ring-2*28fix : 76.491 75.849 76.040 -> 76.491 -> 4283.490 MByte/s p05 ring-1*56fix : 70.131 69.714 70.682 -> 70.682 -> 3958.180 MByte/s p06 random-cyc-1dim : 31.507 32.085 31.606 -> 32.085 -> 1796.748 MByte/s p07 random-cyc-1dim : 37.641 37.177 37.557 -> 37.641 -> 2107.881 MByte/s p08 random-cyc-1dim : 37.148 37.267 37.642 -> 37.642 -> 2107.948 MByte/s p09 random-cyc-1dim : 39.903 40.077 39.676 -> 40.077 -> 2244.329 MByte/s p10 random-cyc-1dim : 46.350 46.684 45.876 -> 46.684 -> 2614.332 MByte/s p11 random-cyc-1dim : 42.382 42.337 42.209 -> 42.382 -> 2373.413 MByte/s p12 random-cyc-1dim : 35.068 34.468 34.972 -> 35.068 -> 1963.811 MByte/s p13 random-cyc-1dim : 36.115 35.768 36.848 -> 36.848 -> 2063.494 MByte/s p14 random-cyc-1dim : 43.404 43.002 43.129 -> 43.404 -> 2430.629 MByte/s p15 random-cyc-1dim : 46.505 46.032 46.752 -> 46.752 -> 2618.112 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 26.200 26.345 26.210 -> 26.345 -> 1475.313 MByte/s p17 best bi-section : 70.821 71.136 70.518 -> 71.136 -> 3983.599 MByte/s p18 worst bi-section : 20.988 20.831 20.773 -> 20.988 -> 1175.310 MByte/s p19 acyclic-1dim-all : 65.214 65.424 65.877 -> 65.877 -> 3689.092 MByte/s p20 acyclic-2dim-all : 52.054 51.965 51.782 -> 52.054 -> 2915.046 MByte/s p21 acyclic-3dim-all : 51.868 52.511 52.350 -> 52.511 -> 2940.640 MByte/s p22 cyclic-1dim-all : 70.911 68.583 69.459 -> 70.911 -> 3971.003 MByte/s p23 cyclic-2dim-all : 53.893 54.177 54.231 -> 54.231 -> 3036.910 MByte/s p24 cyclic-3dim-all : 55.992 55.442 55.706 -> 55.992 -> 3135.563 MByte/s log_avg of all rings : 70.800 70.828 70.421 || 71.091 -> 3981.086 MByte/s log_avg of all random : 39.318 39.212 39.358 || 39.587 -> 2216.881 MByte/s log_avg(ring,random) : 52.761 52.700 52.647 ||( 53.050 -> 2970.790)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-28*2fix p00 method 0 : 0.055 0.871 10.065 51.117 179.780 207.906 -> 71.994 -> 4031.667 MByte/s p00 method 1 : 0.018 0.315 4.851 57.562 181.539 208.448 -> 71.373 -> 3996.887 MByte/s p00 method 2 : 0.025 0.399 5.340 26.099 169.266 162.448 -> 60.302 -> 3376.931 MByte/s p01 ring-14*4fix p01 method 0 : 0.054 0.848 9.944 49.477 148.330 153.774 -> 59.408 -> 3326.838 MByte/s p01 method 1 : 0.031 0.531 7.979 81.404 185.171 192.675 -> 75.798 -> 4244.714 MByte/s p01 method 2 : 0.022 0.345 4.660 29.566 117.946 105.681 -> 45.293 -> 2536.409 MByte/s p02 ring-7*8fix p02 method 0 : 0.052 0.818 9.626 47.909 133.922 138.139 -> 53.409 -> 2990.928 MByte/s p02 method 1 : 0.031 0.531 7.958 77.830 151.037 186.819 -> 67.812 -> 3797.461 MByte/s p02 method 2 : 0.022 0.341 4.487 30.042 118.144 119.090 -> 44.683 -> 2502.268 MByte/s p03 ring-3*18&+1 p03 method 0 : 0.049 0.800 8.517 40.051 136.032 132.621 -> 52.811 -> 2957.413 MByte/s p03 method 1 : 0.031 0.526 7.878 78.477 150.300 163.647 -> 64.998 -> 3639.863 MByte/s p03 method 2 : 0.022 0.342 4.470 29.794 118.858 107.408 -> 44.702 -> 2503.293 MByte/s p04 ring-2*28fix p04 method 0 : 0.048 0.771 8.884 39.016 140.743 131.955 -> 54.433 -> 3048.246 MByte/s p04 method 1 : 0.031 0.531 7.946 80.629 190.285 202.204 -> 77.102 -> 4317.714 MByte/s p04 method 2 : 0.021 0.343 4.491 30.003 126.042 126.092 -> 47.505 -> 2660.275 MByte/s p05 ring-1*56fix p05 method 0 : 0.046 0.755 8.455 38.519 139.413 132.143 -> 53.610 -> 3002.168 MByte/s p05 method 1 : 0.030 0.518 7.835 81.519 164.278 187.205 -> 71.161 -> 3985.032 MByte/s p05 method 2 : 0.022 0.340 4.448 29.838 120.664 96.029 -> 44.699 -> 2503.153 MByte/s p06 random-cyc-1dim p06 method 0 : 0.044 0.701 7.655 35.814 65.222 58.542 -> 29.077 -> 1628.298 MByte/s p06 method 1 : 0.030 0.515 7.789 58.983 64.525 48.642 -> 30.700 -> 1719.182 MByte/s p06 method 2 : 0.021 0.327 4.293 28.949 65.290 52.738 -> 26.841 -> 1503.117 MByte/s p07 random-cyc-1dim p07 method 0 : 0.044 0.713 7.753 37.201 81.603 68.835 -> 33.794 -> 1892.462 MByte/s p07 method 1 : 0.030 0.523 7.837 65.850 81.244 60.829 -> 36.941 -> 2068.710 MByte/s p07 method 2 : 0.021 0.333 4.310 29.191 78.294 73.613 -> 31.068 -> 1739.800 MByte/s p08 random-cyc-1dim p08 method 0 : 0.044 0.712 7.707 37.103 82.667 62.394 -> 33.980 -> 1902.907 MByte/s p08 method 1 : 0.030 0.520 7.785 66.291 77.879 57.811 -> 35.855 -> 2007.879 MByte/s p08 method 2 : 0.021 0.335 4.371 29.242 77.329 69.402 -> 30.519 -> 1709.088 MByte/s p09 random-cyc-1dim p09 method 0 : 0.045 0.703 7.699 37.050 86.986 78.853 -> 35.792 -> 2004.333 MByte/s p09 method 1 : 0.031 0.522 7.827 67.278 86.835 75.076 -> 40.129 -> 2247.243 MByte/s p09 method 2 : 0.021 0.338 4.331 29.256 80.663 69.314 -> 32.037 -> 1794.091 MByte/s p10 random-cyc-1dim p10 method 0 : 0.045 0.697 7.778 37.943 102.008 93.265 -> 41.534 -> 2325.900 MByte/s p10 method 1 : 0.031 0.527 7.906 73.892 103.556 90.433 -> 44.902 -> 2514.534 MByte/s p10 method 2 : 0.021 0.336 4.344 29.418 89.817 79.263 -> 35.157 -> 1968.816 MByte/s p11 random-cyc-1dim p11 method 0 : 0.045 0.710 7.739 36.978 96.343 80.559 -> 38.443 -> 2152.835 MByte/s p11 method 1 : 0.031 0.526 7.832 68.457 91.301 68.810 -> 40.610 -> 2274.171 MByte/s p11 method 2 : 0.021 0.334 4.286 29.196 83.469 78.759 -> 33.414 -> 1871.201 MByte/s p12 random-cyc-1dim p12 method 0 : 0.043 0.700 7.863 37.495 77.320 53.338 -> 31.275 -> 1751.395 MByte/s p12 method 1 : 0.031 0.521 7.770 63.513 70.724 59.246 -> 33.842 -> 1895.151 MByte/s p12 method 2 : 0.021 0.329 4.351 29.140 69.042 66.823 -> 29.052 -> 1626.936 MByte/s p13 random-cyc-1dim p13 method 0 : 0.044 0.690 7.725 36.380 82.830 65.706 -> 33.294 -> 1864.484 MByte/s p13 method 1 : 0.031 0.523 7.812 67.192 71.993 46.838 -> 33.660 -> 1884.943 MByte/s p13 method 2 : 0.021 0.333 4.339 29.068 73.132 68.595 -> 29.725 -> 1664.602 MByte/s p14 random-cyc-1dim p14 method 0 : 0.045 0.738 7.882 37.031 98.276 77.141 -> 38.750 -> 2169.985 MByte/s p14 method 1 : 0.031 0.523 7.849 71.031 94.136 80.100 -> 42.654 -> 2388.647 MByte/s p14 method 2 : 0.021 0.339 4.345 29.347 88.733 77.048 -> 34.807 -> 1949.194 MByte/s p15 random-cyc-1dim p15 method 0 : 0.045 0.707 7.759 37.684 100.995 97.763 -> 40.799 -> 2284.766 MByte/s p15 method 1 : 0.031 0.521 7.811 74.465 111.977 92.924 -> 47.135 -> 2639.583 MByte/s p15 method 2 : 0.021 0.334 4.399 29.562 93.911 81.852 -> 36.351 -> 2035.634 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim p16 method 0 : 0.043 0.672 7.687 33.698 51.767 51.517 -> 24.400 -> 1366.399 MByte/s p16 method 1 : 0.031 0.521 7.841 48.028 50.364 44.112 -> 25.444 -> 1424.839 MByte/s p16 method 2 : 0.021 0.327 4.186 27.839 47.894 44.862 -> 21.645 -> 1212.093 MByte/s p17 best bi-section p17 method 0 : 0.033 0.497 5.496 27.923 131.548 161.089 -> 51.542 -> 2886.379 MByte/s p17 method 1 : 0.019 0.313 4.808 57.257 180.748 206.835 -> 71.281 -> 3991.759 MByte/s p17 method 2 : 0.015 0.236 3.374 27.391 125.288 155.334 -> 49.003 -> 2744.187 MByte/s p18 worst bi-section p18 method 0 : 0.024 0.372 4.506 25.085 37.944 40.230 -> 18.185 -> 1018.364 MByte/s p18 method 1 : 0.018 0.304 4.655 33.580 39.364 56.164 -> 20.980 -> 1174.886 MByte/s p18 method 2 : 0.014 0.229 3.409 24.021 37.475 34.575 -> 17.040 -> 954.215 MByte/s p19 acyclic-1dim-all p19 method 0 : 0.047 0.748 8.457 40.345 141.373 135.235 -> 54.069 -> 3027.873 MByte/s p19 method 1 : 0.031 0.526 7.847 80.102 167.089 145.120 -> 66.989 -> 3751.368 MByte/s p19 method 2 : 0.021 0.337 4.395 29.765 122.790 103.510 -> 45.443 -> 2544.819 MByte/s p20 acyclic-2dim-all p20 method 0 : 0.035 0.552 6.033 28.553 110.705 112.514 -> 43.146 -> 2416.164 MByte/s p20 method 1 : 0.042 0.732 10.490 77.861 114.176 111.807 -> 51.589 -> 2888.983 MByte/s p20 method 2 : 0.019 0.298 3.822 26.241 95.423 86.297 -> 36.521 -> 2045.167 MByte/s p21 acyclic-3dim-all p21 method 0 : 0.032 0.505 5.672 26.208 108.348 118.793 -> 42.137 -> 2359.659 MByte/s p21 method 1 : 0.046 0.802 11.299 81.087 110.098 91.637 -> 50.378 -> 2821.188 MByte/s p21 method 2 : 0.017 0.262 3.471 23.935 100.064 85.889 -> 37.208 -> 2083.648 MByte/s p22 cyclic-1dim-all p22 method 0 : 0.047 0.749 8.285 39.551 139.777 132.687 -> 54.031 -> 3025.714 MByte/s p22 method 1 : 0.030 0.505 7.638 78.628 169.305 186.287 -> 71.158 -> 3984.864 MByte/s p22 method 2 : 0.022 0.346 4.475 29.869 120.780 107.279 -> 44.811 -> 2509.425 MByte/s p23 cyclic-2dim-all p23 method 0 : 0.040 0.623 6.656 32.692 121.157 122.222 -> 46.812 -> 2621.465 MByte/s p23 method 1 : 0.045 0.788 11.414 79.214 111.284 92.837 -> 50.197 -> 2811.039 MByte/s p23 method 2 : 0.022 0.335 4.395 29.622 100.939 86.512 -> 38.749 -> 2169.952 MByte/s p24 cyclic-3dim-all p24 method 0 : 0.040 0.620 6.748 32.700 118.043 126.358 -> 46.583 -> 2608.660 MByte/s p24 method 1 : 0.050 0.887 12.554 84.163 116.447 108.830 -> 53.472 -> 2994.457 MByte/s p24 method 2 : 0.022 0.343 4.439 29.714 109.311 95.101 -> 41.583 -> 2328.664 MByte/s log_avg of all rings - ring, method 0 : 0.051 0.810 9.225 44.037 145.613 147.296 || 57.249 -> 3205.924 MByte/s - ring, method 1 : 0.028 0.484 7.298 75.696 169.666 189.617 || 71.250 -> 3989.995 MByte/s - ring, method 2 : 0.022 0.351 4.639 29.188 127.340 117.702 || 47.568 -> 2663.791 MByte/s log_avg of all random - random, method 0 : 0.044 0.707 7.756 37.063 86.668 72.363 || 35.458 -> 1985.634 MByte/s - random, method 1 : 0.030 0.522 7.822 67.549 84.237 66.338 || 38.313 -> 2145.517 MByte/s - random, method 2 : 0.021 0.334 4.337 29.237 79.478 71.247 || 31.767 -> 1778.960 MByte/s log_avg(ring,random) - average, method 0 : 0.047 0.757 8.459 40.400 112.338 103.242 || 45.054 -> 2523.052 MByte/s - average, method 1 : 0.029 0.503 7.555 71.507 119.550 112.155 || 52.247 -> 2925.851 MByte/s - average, method 2 : 0.022 0.342 4.486 29.212 100.602 91.575 || 38.873 -> 2176.874 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 2.653 0.047 0.051 0.044 0.047 0.029 0.022 2 5.325 0.095 0.102 0.089 0.095 0.058 0.043 4 10.631 0.190 0.203 0.178 0.190 0.128 0.087 8 21.224 0.379 0.406 0.354 0.379 0.255 0.173 16 42.370 0.757 0.810 0.707 0.757 0.503 0.342 32 64.976 1.160 1.279 1.053 1.160 0.999 0.598 64 119.496 2.134 2.308 1.973 2.107 1.908 1.144 128 257.534 4.599 5.029 4.206 4.599 3.951 2.368 256 476.071 8.501 9.225 7.834 8.459 7.555 4.486 512 869.353 15.524 16.205 14.872 15.056 14.421 8.184 1024 1563.223 27.915 28.308 27.527 23.116 26.907 14.101 2048 2626.704 46.905 47.454 46.364 31.568 46.357 21.693 4096 4004.367 71.507 75.696 67.549 40.400 71.507 29.212 8192 5196.928 92.802 108.411 79.441 64.516 92.802 61.559 16384 6054.685 108.119 139.207 83.974 87.006 108.119 81.014 32768 6519.593 116.421 159.326 85.071 102.973 116.272 93.260 65536 6831.319 121.988 169.666 87.708 112.338 119.550 100.602 131072 6973.293 124.523 177.414 87.400 116.600 119.470 101.472 262144 6994.994 124.911 183.175 85.179 115.357 119.740 100.990 524288 6971.886 124.498 188.439 82.253 110.967 117.155 99.811 1048576 6718.923 119.981 189.617 75.918 103.242 112.155 91.575 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-28*2fix : 0.055 0.871 10.065 57.562 181.539 208.448 -> 73.038 -> 4090.137 MByte/s p01 ring-14*4fix : 0.054 0.848 9.944 81.404 185.171 192.675 -> 76.115 -> 4262.421 MByte/s p02 ring-7*8fix : 0.052 0.818 9.626 77.830 151.037 186.819 -> 68.053 -> 3810.966 MByte/s p03 ring-3*18&+1 : 0.049 0.800 8.517 78.477 150.300 163.647 -> 65.124 -> 3646.929 MByte/s p04 ring-2*28fix : 0.048 0.771 8.884 80.629 190.285 202.204 -> 77.237 -> 4325.260 MByte/s p05 ring-1*56fix : 0.046 0.755 8.455 81.519 164.278 187.205 -> 71.244 -> 3989.661 MByte/s p06 random-cyc-1dim : 0.044 0.701 7.789 58.983 65.290 58.542 -> 32.264 -> 1806.792 MByte/s p07 random-cyc-1dim : 0.044 0.713 7.837 65.850 81.603 73.613 -> 38.081 -> 2132.516 MByte/s p08 random-cyc-1dim : 0.044 0.712 7.785 66.291 82.667 69.402 -> 38.085 -> 2132.777 MByte/s p09 random-cyc-1dim : 0.045 0.703 7.827 67.278 86.986 78.853 -> 40.338 -> 2258.940 MByte/s p10 random-cyc-1dim : 0.045 0.697 7.906 73.892 103.556 93.265 -> 46.864 -> 2624.398 MByte/s p11 random-cyc-1dim : 0.045 0.710 7.832 68.457 96.343 80.559 -> 43.148 -> 2416.265 MByte/s p12 random-cyc-1dim : 0.043 0.700 7.863 63.513 77.320 66.823 -> 35.536 -> 1990.015 MByte/s p13 random-cyc-1dim : 0.044 0.690 7.812 67.192 82.830 68.595 -> 37.389 -> 2093.788 MByte/s p14 random-cyc-1dim : 0.045 0.738 7.882 71.031 98.276 80.100 -> 43.860 -> 2456.135 MByte/s p15 random-cyc-1dim : 0.045 0.707 7.811 74.465 111.977 97.763 -> 47.609 -> 2666.131 MByte/s -- additional patterns that are not used to compute b_eff: p16 worst-cyc-1dim : 0.043 0.672 7.841 48.028 51.767 51.517 -> 26.691 -> 1494.713 MByte/s p17 best bi-section : 0.033 0.497 5.496 57.257 180.748 206.835 -> 71.395 -> 3998.122 MByte/s p18 worst bi-section : 0.024 0.372 4.655 33.580 39.364 56.164 -> 21.023 -> 1177.308 MByte/s p19 acyclic-1dim-all : 0.047 0.748 8.457 80.102 167.089 145.120 -> 67.071 -> 3755.985 MByte/s p20 acyclic-2dim-all : 0.042 0.732 10.490 77.861 114.176 112.514 -> 52.632 -> 2947.374 MByte/s p21 acyclic-3dim-all : 0.046 0.802 11.299 81.087 110.098 118.793 -> 52.846 -> 2959.370 MByte/s p22 cyclic-1dim-all : 0.047 0.749 8.285 78.628 169.305 186.287 -> 71.250 -> 3990.027 MByte/s p23 cyclic-2dim-all : 0.045 0.788 11.414 79.214 121.157 122.222 -> 54.828 -> 3070.367 MByte/s p24 cyclic-3dim-all : 0.050 0.887 12.554 84.163 118.043 126.358 -> 56.162 -> 3145.078 MByte/s log_avg of all rings : 0.051 0.810 9.225 75.696 169.666 189.617 || 71.674 -> 4013.754 MByte/s log_avg of all random : 0.044 0.707 7.834 67.549 87.708 75.918 || 40.039 -> 2242.202 MByte/s log_avg(ring,random) : 0.047 0.757 8.501 71.507 121.988 119.981 || 53.570 -> 2999.941 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 2999.941 MByte/s on 56 processes ( = 53.570 MByte/s * 56 processes) system parameters : 56 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 = 2999.941 MB/s = 53.570 * 56 PEs with 128 MB/PE on sn6715 hwwt3e 2.0.4.71 unicosmk CRAY T3E