SECTION-HEAD-BEGIN b_eff.c, Revision 3.5 from March 27, 2002 MEMORY_PER_PROCESSOR = 1024 MBytes [1M = 1024*1024] 2-dim-patterns: size = 12 * 8 3-dim-patterns: size = 6 * 4 * 4 Used message lengths: msglng= 1 2 4 8 16 32 64 128 256 512 1024 2048 4096 10624 27554 71468 185364 480774 1246974 3234251 8388608 Used methods: 0=Sndrcv 1=Alltoal 2=non-blk Used patterns: 0=ring-48*2fix 1=ring-24*4fix 2=ring-12*8fix 3=ring-4*24fix 4=ring-2*48fix 5=ring-1*96fix 6=random-cyc-1dim 7=random-cyc-1dim 8=random-cyc-1dim 9=random-cyc-1dim 10=random-cyc-1dim 11=random-cyc-1dim 12=random-cyc-1dim 13=random-cyc-1dim 14=random-cyc-1dim 15=random-cyc-1dim 16=random-cyc-1dim 17=random-cyc-1dim 18=random-cyc-1dim 19=random-cyc-1dim 20=random-cyc-1dim 21=random-cyc-1dim 22=random-cyc-1dim 23=random-cyc-1dim 24=random-cyc-1dim 25=random-cyc-1dim 26=random-cyc-1dim 27=random-cyc-1dim 28=random-cyc-1dim 29=random-cyc-1dim 30=random-cyc-1dim 31=random-cyc-1dim 32=random-cyc-1dim 33=random-cyc-1dim 34=random-cyc-1dim 35=random-cyc-1dim 36=worst-cyc-1dim 37=best bi-section 38=worst bi-section 39=one PingPong Pair 40=acyclic-2dim-all 41=acyclic-3dim-all 42=cyclic-2dim-x 43=cyclic-2dim-y 44=cyclic-2dim-all 45=cyclic-3dim-x 46=cyclic-3dim-y 47=cyclic-3dim-z 48=cyclic-3dim-all 0-5 used for ring pattern average of b_eff 6-35 used for random pattern average of b_eff 36-47 only reported, not used for b_eff average SECTION-HEAD-END SECTION-ELAPSED-BEGIN measurment loop: elapsed time = 159.244 sec sum of max elapsed time per entries above = 152.794 sec difference to elapsed time = 6.450 sec = 4.1% sum based on fastest repetition = 153.717 sec difference to elapsed time = 5.526 sec = 3.5% The difference is less than 5 % SECTION-ELAPSED-END SECTION-PATTERN-BEGIN Pattern parameters: sendrecv total number messages/ messages/ unused best method calls of messages used node node&call nodes per repetition ---- ----------------- ----- ------------ --------- --------- ------ -------------- p00 ring-48*2fix 1 96 1.00 1.00 0 ( 2 0 0 ) p01 ring-24*4fix 2 192 2.00 1.00 0 ( 0 2 2 ) p02 ring-12*8fix 2 192 2.00 1.00 0 ( 0 0 0 ) p03 ring-4*24fix 2 192 2.00 1.00 0 ( 2 0 2 ) p04 ring-2*48fix 2 192 2.00 1.00 0 ( 2 0 0 ) p05 ring-1*96fix 2 192 2.00 1.00 0 ( 2 0 0 ) p06 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p07 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p08 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 2 0 ) p09 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 2 2 ) p10 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p11 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 2 2 ) p12 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p13 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p14 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p15 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p16 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p17 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 2 2 ) p18 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p19 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 2 ) p20 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p21 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p22 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 2 2 ) p23 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 2 2 ) p24 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 2 ) p25 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p26 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p27 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p28 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p29 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p30 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p31 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p32 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p33 random-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 0 ) p34 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p35 random-cyc-1dim 2 192 2.00 1.00 0 ( 0 0 0 ) p36 worst-cyc-1dim 2 192 2.00 1.00 0 ( 2 0 2 ) p37 best bi-section 2 96 1.00 0.50 0 ( 2 2 2 ) p38 worst bi-section 2 96 1.00 0.50 0 ( 2 1 2 ) p39 one PingPong Pair 2 2 1.00 0.50 94 ( 0 0 0 ) p40 acyclic-2dim-all 4 344 3.58 0.90 0 ( 0 0 0 ) p41 acyclic-3dim-all 6 448 4.67 0.78 0 ( 0 0 0 ) p42 cyclic-2dim-x 2 192 2.00 1.00 0 ( 0 0 0 ) p43 cyclic-2dim-y 2 192 2.00 1.00 0 ( 2 0 2 ) p44 cyclic-2dim-all 4 384 4.00 1.00 0 ( 2 2 0 ) p45 cyclic-3dim-x 2 192 2.00 1.00 0 ( 0 0 0 ) p46 cyclic-3dim-y 2 192 2.00 1.00 0 ( 0 2 0 ) p47 cyclic-3dim-z 2 192 2.00 1.00 0 ( 0 2 0 ) p48 cyclic-3dim-all 6 576 6.00 1.00 0 ( 0 0 0 ) 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-48*2fix : 177.512 62.921 160.709 -> 177.512 -> 17041.176 MByte/s p01 ring-24*4fix : 174.216 64.314 160.915 -> 174.216 -> 16724.709 MByte/s p02 ring-12*8fix : 169.558 63.008 157.916 -> 169.558 -> 16277.604 MByte/s p03 ring-4*24fix : 108.753 44.308 107.581 -> 108.753 -> 10440.276 MByte/s p04 ring-2*48fix : 112.873 41.272 107.334 -> 112.873 -> 10835.799 MByte/s p05 ring-1*96fix : 119.139 44.045 107.324 -> 119.139 -> 11437.350 MByte/s p06 random-cyc-1dim : 28.753 11.247 25.361 -> 28.753 -> 2760.324 MByte/s p07 random-cyc-1dim : 27.610 10.996 24.520 -> 27.610 -> 2650.516 MByte/s p08 random-cyc-1dim : 26.252 10.307 23.244 -> 26.252 -> 2520.217 MByte/s p09 random-cyc-1dim : 27.032 10.448 23.798 -> 27.032 -> 2595.063 MByte/s p10 random-cyc-1dim : 28.648 11.455 26.030 -> 28.648 -> 2750.185 MByte/s p11 random-cyc-1dim : 27.550 10.352 24.374 -> 27.550 -> 2644.791 MByte/s p12 random-cyc-1dim : 26.897 10.871 24.586 -> 26.897 -> 2582.157 MByte/s p13 random-cyc-1dim : 27.688 10.451 25.096 -> 27.688 -> 2658.081 MByte/s p14 random-cyc-1dim : 30.406 12.114 28.070 -> 30.406 -> 2919.008 MByte/s p15 random-cyc-1dim : 30.203 11.732 27.279 -> 30.203 -> 2899.497 MByte/s p16 random-cyc-1dim : 28.360 11.396 23.958 -> 28.360 -> 2722.594 MByte/s p17 random-cyc-1dim : 26.724 9.652 23.440 -> 26.724 -> 2565.538 MByte/s p18 random-cyc-1dim : 28.022 10.781 24.089 -> 28.022 -> 2690.123 MByte/s p19 random-cyc-1dim : 28.457 11.223 23.721 -> 28.457 -> 2731.837 MByte/s p20 random-cyc-1dim : 26.203 10.714 22.687 -> 26.203 -> 2515.484 MByte/s p21 random-cyc-1dim : 27.775 11.109 24.732 -> 27.775 -> 2666.440 MByte/s p22 random-cyc-1dim : 27.160 10.102 24.184 -> 27.160 -> 2607.346 MByte/s p23 random-cyc-1dim : 26.953 9.750 22.921 -> 26.953 -> 2587.481 MByte/s p24 random-cyc-1dim : 27.865 10.964 24.750 -> 27.865 -> 2675.031 MByte/s p25 random-cyc-1dim : 27.800 11.200 24.746 -> 27.800 -> 2668.828 MByte/s p26 random-cyc-1dim : 28.924 11.510 25.705 -> 28.924 -> 2776.698 MByte/s p27 random-cyc-1dim : 27.617 11.179 25.746 -> 27.617 -> 2651.238 MByte/s p28 random-cyc-1dim : 29.889 11.803 26.431 -> 29.889 -> 2869.322 MByte/s p29 random-cyc-1dim : 28.376 11.398 25.821 -> 28.376 -> 2724.115 MByte/s p30 random-cyc-1dim : 28.141 11.154 24.715 -> 28.141 -> 2701.531 MByte/s p31 random-cyc-1dim : 27.596 11.128 23.686 -> 27.596 -> 2649.222 MByte/s p32 random-cyc-1dim : 28.155 11.153 24.226 -> 28.155 -> 2702.925 MByte/s p33 random-cyc-1dim : 28.326 11.171 25.933 -> 28.326 -> 2719.307 MByte/s p34 random-cyc-1dim : 27.548 11.000 23.630 -> 27.548 -> 2644.563 MByte/s p35 random-cyc-1dim : 27.483 10.712 25.141 -> 27.483 -> 2638.345 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 34.816 12.685 28.652 -> 34.816 -> 3342.349 MByte/s p37 best bi-section : 165.703 63.349 165.959 -> 165.959 -> 15932.027 MByte/s p38 worst bi-section : 27.265 19.356 31.087 -> 31.087 -> 2984.359 MByte/s p39 one PingPong Pair : 3.759 1.256 1.256 -> 3.759 -> 360.903 MByte/s p40 acyclic-2dim-all : 47.659 21.736 43.214 -> 47.659 -> 4575.282 MByte/s p41 acyclic-3dim-all : 48.128 23.189 44.671 -> 48.128 -> 4620.308 MByte/s p42 cyclic-2dim-x : 31.363 13.555 27.558 -> 31.363 -> 3010.846 MByte/s p43 cyclic-2dim-y : 171.155 62.913 156.984 -> 171.155 -> 16430.907 MByte/s p44 cyclic-2dim-all : 52.101 21.786 45.344 -> 52.101 -> 5001.687 MByte/s p45 cyclic-3dim-x : 31.872 13.525 26.445 -> 31.872 -> 3059.699 MByte/s p46 cyclic-3dim-y : 67.103 26.406 63.975 -> 67.103 -> 6441.847 MByte/s p47 cyclic-3dim-z : 175.182 64.559 162.814 -> 175.182 -> 16817.489 MByte/s p48 cyclic-3dim-all : 58.472 28.245 49.007 -> 58.472 -> 5613.299 MByte/s log_avg of all rings : 140.428 52.331 131.030 || 140.428 -> 13481.074 MByte/s log_avg of all random : 27.930 10.954 24.724 || 27.930 -> 2681.245 MByte/s log_avg(ring,random) : 62.627 23.943 56.918 ||( 62.627 -> 6012.159)MByte/s * size -> accumulated on all pr.: 6012.159 2298.491 5464.118 ||(6012.159)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-48*2fix : 170.808 166.313 171.136 -> 171.136 -> 16429.025 MByte/s p01 ring-24*4fix : 168.665 165.880 170.882 -> 170.882 -> 16404.649 MByte/s p02 ring-12*8fix : 168.265 167.919 163.691 -> 168.265 -> 16153.422 MByte/s p03 ring-4*24fix : 87.261 107.903 107.933 -> 107.933 -> 10361.559 MByte/s p04 ring-2*48fix : 81.820 110.252 110.248 -> 110.252 -> 10584.189 MByte/s p05 ring-1*96fix : 92.915 115.065 114.388 -> 115.065 -> 11046.283 MByte/s p06 random-cyc-1dim : 24.421 26.992 28.122 -> 28.122 -> 2699.739 MByte/s p07 random-cyc-1dim : 24.334 25.408 27.339 -> 27.339 -> 2624.547 MByte/s p08 random-cyc-1dim : 23.092 23.969 26.198 -> 26.198 -> 2514.974 MByte/s p09 random-cyc-1dim : 23.442 24.645 26.375 -> 26.375 -> 2531.990 MByte/s p10 random-cyc-1dim : 24.980 27.325 28.354 -> 28.354 -> 2721.954 MByte/s p11 random-cyc-1dim : 24.378 26.575 26.134 -> 26.575 -> 2551.158 MByte/s p12 random-cyc-1dim : 23.327 26.074 26.536 -> 26.536 -> 2547.429 MByte/s p13 random-cyc-1dim : 23.271 26.906 26.198 -> 26.906 -> 2582.931 MByte/s p14 random-cyc-1dim : 24.788 29.630 29.499 -> 29.630 -> 2844.462 MByte/s p15 random-cyc-1dim : 24.729 28.997 29.983 -> 29.983 -> 2878.343 MByte/s p16 random-cyc-1dim : 24.383 26.967 28.326 -> 28.326 -> 2719.271 MByte/s p17 random-cyc-1dim : 23.533 25.998 25.677 -> 25.998 -> 2495.776 MByte/s p18 random-cyc-1dim : 24.276 26.960 27.985 -> 27.985 -> 2686.582 MByte/s p19 random-cyc-1dim : 23.359 27.918 26.791 -> 27.918 -> 2680.118 MByte/s p20 random-cyc-1dim : 23.481 25.320 26.083 -> 26.083 -> 2503.935 MByte/s p21 random-cyc-1dim : 23.648 26.458 26.450 -> 26.458 -> 2540.001 MByte/s p22 random-cyc-1dim : 23.992 25.734 26.394 -> 26.394 -> 2533.852 MByte/s p23 random-cyc-1dim : 24.224 25.587 26.794 -> 26.794 -> 2572.238 MByte/s p24 random-cyc-1dim : 24.154 26.855 26.526 -> 26.855 -> 2578.103 MByte/s p25 random-cyc-1dim : 24.605 27.310 27.174 -> 27.310 -> 2621.725 MByte/s p26 random-cyc-1dim : 25.154 28.307 28.127 -> 28.307 -> 2717.499 MByte/s p27 random-cyc-1dim : 23.623 26.350 26.907 -> 26.907 -> 2583.082 MByte/s p28 random-cyc-1dim : 26.035 28.585 29.360 -> 29.360 -> 2818.602 MByte/s p29 random-cyc-1dim : 25.293 27.224 28.438 -> 28.438 -> 2730.059 MByte/s p30 random-cyc-1dim : 25.081 26.550 28.211 -> 28.211 -> 2708.282 MByte/s p31 random-cyc-1dim : 23.395 26.798 27.177 -> 27.177 -> 2608.986 MByte/s p32 random-cyc-1dim : 23.732 27.122 27.841 -> 27.841 -> 2672.709 MByte/s p33 random-cyc-1dim : 25.106 27.055 28.292 -> 28.292 -> 2716.067 MByte/s p34 random-cyc-1dim : 24.775 26.705 27.290 -> 27.290 -> 2619.881 MByte/s p35 random-cyc-1dim : 24.342 26.570 27.278 -> 27.278 -> 2618.716 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 25.863 31.670 32.159 -> 32.159 -> 3087.297 MByte/s p37 best bi-section : 166.577 165.984 167.301 -> 167.301 -> 16060.927 MByte/s p38 worst bi-section : 30.043 32.080 30.945 -> 32.080 -> 3079.674 MByte/s p39 one PingPong Pair : 3.750 3.702 3.699 -> 3.750 -> 359.984 MByte/s p40 acyclic-2dim-all : 41.172 46.101 47.094 -> 47.094 -> 4521.071 MByte/s p41 acyclic-3dim-all : 40.685 47.134 48.438 -> 48.438 -> 4650.031 MByte/s p42 cyclic-2dim-x : 29.834 30.911 30.287 -> 30.911 -> 2967.484 MByte/s p43 cyclic-2dim-y : 168.112 168.208 168.059 -> 168.208 -> 16147.961 MByte/s p44 cyclic-2dim-all : 48.522 49.819 52.047 -> 52.047 -> 4996.487 MByte/s p45 cyclic-3dim-x : 29.501 30.749 31.232 -> 31.232 -> 2998.265 MByte/s p46 cyclic-3dim-y : 50.034 64.531 67.332 -> 67.332 -> 6463.878 MByte/s p47 cyclic-3dim-z : 173.767 170.496 174.119 -> 174.119 -> 16715.403 MByte/s p48 cyclic-3dim-all : 52.078 56.011 57.169 -> 57.169 -> 5488.202 MByte/s log_avg of all rings : 121.492 136.050 136.666 || 137.431 -> 13193.409 MByte/s log_avg of all random : 24.221 26.737 27.374 || 27.489 -> 2638.910 MByte/s log_avg(ring,random) : 54.246 60.312 61.165 ||( 61.464 -> 5900.527)MByte/s * size -> accumulated on all pr.: 5207.662 5789.999 5871.793 ||(5900.527)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 185364 8388608 -> average -> accumulated (latency of one sendrecv (or equivalent method) in microsec) p00 ring-48*2fix p00 method 0 =Sndrcv :( 15.347) 0.065 1.012 13.424 106.607 386.130 635.981 -> 177.512 -> 17041.176 MByte/s p00 method 1 =Alltoal :(3275.585) 0.000 0.005 0.078 1.253 18.245 635.981 -> 62.921 -> 6040.414 MByte/s p00 method 2 =non-blk :( 38.259) 0.026 0.431 6.465 72.038 344.548 635.981 -> 160.709 -> 15428.027 MByte/s p01 ring-24*4fix p01 method 0 =Sndrcv :( 15.537) 0.064 1.018 14.518 108.003 382.615 629.563 -> 174.216 -> 16724.709 MByte/s p01 method 1 =Alltoal :(1629.102) 0.001 0.010 0.155 2.499 26.104 629.563 -> 64.314 -> 6174.136 MByte/s p01 method 2 =non-blk :( 37.193) 0.027 0.433 6.549 73.165 378.701 629.563 -> 160.915 -> 15447.837 MByte/s p02 ring-12*8fix p02 method 0 =Sndrcv :( 15.438) 0.065 1.024 14.489 108.278 358.530 634.396 -> 169.558 -> 16277.604 MByte/s p02 method 1 =Alltoal :(1637.006) 0.001 0.010 0.157 2.492 23.001 634.396 -> 63.008 -> 6048.751 MByte/s p02 method 2 =non-blk :( 37.177) 0.027 0.451 6.595 75.205 371.109 634.396 -> 157.916 -> 15159.950 MByte/s p03 ring-4*24fix p03 method 0 =Sndrcv :( 26.544) 0.038 0.583 8.111 63.900 236.584 451.827 -> 108.753 -> 10440.276 MByte/s p03 method 1 =Alltoal :(1637.805) 0.001 0.010 0.157 2.448 20.846 451.827 -> 44.308 -> 4253.575 MByte/s p03 method 2 =non-blk :( 43.561) 0.023 0.358 5.398 49.555 233.160 451.827 -> 107.581 -> 10327.768 MByte/s p04 ring-2*48fix p04 method 0 =Sndrcv :( 26.725) 0.037 0.583 8.254 64.274 261.270 425.375 -> 112.873 -> 10835.799 MByte/s p04 method 1 =Alltoal :(1635.301) 0.001 0.010 0.156 2.432 20.604 425.375 -> 41.272 -> 3962.144 MByte/s p04 method 2 =non-blk :( 43.061) 0.023 0.363 5.472 50.082 246.172 425.375 -> 107.334 -> 10304.081 MByte/s p05 ring-1*96fix p05 method 0 =Sndrcv :( 26.806) 0.037 0.576 8.032 64.128 287.820 456.077 -> 119.139 -> 11437.350 MByte/s p05 method 1 =Alltoal :(1636.291) 0.001 0.010 0.158 2.414 18.932 456.077 -> 44.045 -> 4228.301 MByte/s p05 method 2 =non-blk :( 42.938) 0.023 0.360 5.495 49.962 231.116 456.077 -> 107.324 -> 10303.078 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 32.869) 0.030 0.477 6.906 34.851 57.097 84.069 -> 28.753 -> 2760.324 MByte/s p06 method 1 =Alltoal :(1613.307) 0.001 0.010 0.160 1.804 14.611 84.069 -> 11.247 -> 1079.671 MByte/s p06 method 2 =non-blk :( 50.704) 0.020 0.301 4.633 31.669 47.719 84.069 -> 25.361 -> 2434.659 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 33.123) 0.030 0.476 6.757 33.795 55.498 80.849 -> 27.610 -> 2650.516 MByte/s p07 method 1 =Alltoal :(1609.492) 0.001 0.010 0.157 1.767 14.393 80.849 -> 10.996 -> 1055.634 MByte/s p07 method 2 =non-blk :( 51.132) 0.020 0.305 4.630 29.574 46.185 80.849 -> 24.520 -> 2353.916 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 32.663) 0.031 0.475 6.845 33.720 47.154 69.754 -> 26.252 -> 2520.217 MByte/s p08 method 1 =Alltoal :(1610.899) 0.001 0.010 0.159 1.752 14.087 69.754 -> 10.307 -> 989.491 MByte/s p08 method 2 =non-blk :( 50.958) 0.020 0.304 4.623 30.843 43.203 69.754 -> 23.244 -> 2231.377 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 32.594) 0.031 0.471 6.678 33.716 54.097 79.607 -> 27.032 -> 2595.063 MByte/s p09 method 1 =Alltoal :(1608.503) 0.001 0.010 0.159 1.730 15.068 79.607 -> 10.448 -> 1002.983 MByte/s p09 method 2 =non-blk :( 50.950) 0.020 0.300 4.633 30.514 41.730 79.607 -> 23.798 -> 2284.654 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 32.612) 0.031 0.473 6.806 34.970 55.045 85.325 -> 28.648 -> 2750.185 MByte/s p10 method 1 =Alltoal :(1609.302) 0.001 0.010 0.160 1.779 15.701 85.325 -> 11.455 -> 1099.656 MByte/s p10 method 2 =non-blk :( 51.520) 0.019 0.302 4.683 32.453 47.991 85.325 -> 26.030 -> 2498.848 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 32.794) 0.030 0.475 6.856 34.166 53.151 78.694 -> 27.550 -> 2644.791 MByte/s p11 method 1 =Alltoal :(1609.898) 0.001 0.010 0.158 1.794 14.232 78.694 -> 10.352 -> 993.824 MByte/s p11 method 2 =non-blk :( 51.296) 0.019 0.307 4.647 31.863 44.548 78.694 -> 24.374 -> 2339.897 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 32.656) 0.031 0.477 6.731 34.076 54.359 81.316 -> 26.897 -> 2582.157 MByte/s p12 method 1 =Alltoal :(1611.996) 0.001 0.010 0.156 1.753 13.866 81.316 -> 10.871 -> 1043.580 MByte/s p12 method 2 =non-blk :( 52.240) 0.019 0.304 4.591 30.417 48.411 81.316 -> 24.586 -> 2360.214 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 32.600) 0.031 0.471 6.764 34.491 54.606 82.141 -> 27.688 -> 2658.081 MByte/s p13 method 1 =Alltoal :(1607.704) 0.001 0.010 0.159 1.747 13.251 82.141 -> 10.451 -> 1003.327 MByte/s p13 method 2 =non-blk :( 51.725) 0.019 0.307 4.652 31.374 46.904 82.141 -> 25.096 -> 2409.215 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 32.350) 0.031 0.476 6.988 39.004 57.772 91.550 -> 30.406 -> 2919.008 MByte/s p14 method 1 =Alltoal :(1611.507) 0.001 0.010 0.160 1.796 14.286 91.550 -> 12.114 -> 1162.913 MByte/s p14 method 2 =non-blk :( 51.511) 0.019 0.302 4.685 33.470 60.846 91.550 -> 28.070 -> 2694.759 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 32.543) 0.031 0.477 6.792 36.806 62.151 87.939 -> 30.203 -> 2899.497 MByte/s p15 method 1 =Alltoal :(1612.699) 0.001 0.010 0.160 1.868 13.017 87.939 -> 11.732 -> 1126.295 MByte/s p15 method 2 =non-blk :( 51.552) 0.019 0.303 4.592 32.923 52.810 87.939 -> 27.279 -> 2618.804 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 32.638) 0.031 0.477 6.885 35.818 55.891 84.847 -> 28.360 -> 2722.594 MByte/s p16 method 1 =Alltoal :(1608.205) 0.001 0.010 0.158 1.782 15.585 84.847 -> 11.396 -> 1093.986 MByte/s p16 method 2 =non-blk :( 51.266) 0.020 0.308 4.587 31.443 35.524 84.847 -> 23.958 -> 2299.973 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 32.593) 0.031 0.473 6.827 34.313 55.054 71.058 -> 26.724 -> 2565.538 MByte/s p17 method 1 =Alltoal :(1609.504) 0.001 0.010 0.156 1.748 14.396 71.058 -> 9.652 -> 926.557 MByte/s p17 method 2 =non-blk :( 51.221) 0.020 0.305 4.624 31.318 45.416 71.058 -> 23.440 -> 2250.198 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 32.593) 0.031 0.471 6.838 34.170 59.081 80.798 -> 28.022 -> 2690.123 MByte/s p18 method 1 =Alltoal :(1612.198) 0.001 0.010 0.160 1.767 12.677 80.798 -> 10.781 -> 1035.004 MByte/s p18 method 2 =non-blk :( 51.459) 0.019 0.302 4.653 31.297 44.430 80.798 -> 24.089 -> 2312.506 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 32.562) 0.031 0.476 6.823 35.328 55.993 82.750 -> 28.457 -> 2731.837 MByte/s p19 method 1 =Alltoal :(1602.805) 0.001 0.010 0.159 1.774 15.618 82.750 -> 11.223 -> 1077.420 MByte/s p19 method 2 =non-blk :( 51.279) 0.020 0.298 4.645 31.829 46.975 82.750 -> 23.721 -> 2277.203 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 32.694) 0.031 0.475 6.622 31.875 49.808 79.735 -> 26.203 -> 2515.484 MByte/s p20 method 1 =Alltoal :(1601.899) 0.001 0.010 0.158 1.708 12.380 79.735 -> 10.714 -> 1028.501 MByte/s p20 method 2 =non-blk :( 51.861) 0.019 0.304 4.540 26.232 41.890 79.735 -> 22.687 -> 2177.915 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 32.746) 0.031 0.474 6.904 34.154 53.565 83.089 -> 27.775 -> 2666.440 MByte/s p21 method 1 =Alltoal :(1601.696) 0.001 0.010 0.160 1.736 13.365 83.089 -> 11.109 -> 1066.471 MByte/s p21 method 2 =non-blk :( 51.330) 0.019 0.305 4.583 31.625 44.250 83.089 -> 24.732 -> 2374.316 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 32.593) 0.031 0.472 6.851 34.293 59.459 74.402 -> 27.160 -> 2607.346 MByte/s p22 method 1 =Alltoal :(1606.703) 0.001 0.010 0.157 1.753 15.398 74.402 -> 10.102 -> 969.757 MByte/s p22 method 2 =non-blk :( 50.582) 0.020 0.303 4.589 31.267 54.631 74.402 -> 24.184 -> 2321.649 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 32.549) 0.031 0.474 6.910 35.507 55.102 73.939 -> 26.953 -> 2587.481 MByte/s p23 method 1 =Alltoal :(1606.989) 0.001 0.010 0.160 1.790 13.236 73.939 -> 9.750 -> 935.996 MByte/s p23 method 2 =non-blk :( 50.786) 0.020 0.303 4.651 31.840 40.897 73.939 -> 22.921 -> 2200.431 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 32.649) 0.031 0.477 6.748 34.060 57.701 79.343 -> 27.865 -> 2675.031 MByte/s p24 method 1 =Alltoal :(1604.295) 0.001 0.010 0.157 1.757 14.511 79.343 -> 10.964 -> 1052.501 MByte/s p24 method 2 =non-blk :( 50.805) 0.020 0.302 4.635 30.880 48.493 79.343 -> 24.750 -> 2375.989 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 33.049) 0.030 0.475 6.751 33.823 57.908 82.913 -> 27.800 -> 2668.828 MByte/s p25 method 1 =Alltoal :(1604.998) 0.001 0.010 0.158 1.717 15.035 82.913 -> 11.200 -> 1075.189 MByte/s p25 method 2 =non-blk :( 50.960) 0.020 0.306 4.542 30.938 45.233 82.913 -> 24.746 -> 2375.629 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 32.613) 0.031 0.471 6.980 35.206 58.318 84.835 -> 28.924 -> 2776.698 MByte/s p26 method 1 =Alltoal :(1605.201) 0.001 0.010 0.160 1.820 14.521 84.835 -> 11.510 -> 1104.932 MByte/s p26 method 2 =non-blk :( 50.337) 0.020 0.308 4.617 30.369 49.097 84.835 -> 25.705 -> 2467.720 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 32.619) 0.031 0.472 6.781 34.611 52.430 83.641 -> 27.617 -> 2651.238 MByte/s p27 method 1 =Alltoal :(1609.397) 0.001 0.010 0.157 1.700 14.628 83.641 -> 11.179 -> 1073.161 MByte/s p27 method 2 =non-blk :( 50.929) 0.020 0.301 4.567 30.863 49.849 83.641 -> 25.746 -> 2471.612 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 32.607) 0.031 0.478 6.935 37.188 57.647 89.477 -> 29.889 -> 2869.322 MByte/s p28 method 1 =Alltoal :(1612.508) 0.001 0.010 0.160 1.741 14.545 89.477 -> 11.803 -> 1133.048 MByte/s p28 method 2 =non-blk :( 50.460) 0.020 0.302 4.652 32.637 48.928 89.477 -> 26.431 -> 2537.348 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 32.581) 0.031 0.477 6.821 35.034 47.830 84.770 -> 28.376 -> 2724.115 MByte/s p29 method 1 =Alltoal :(1610.804) 0.001 0.010 0.159 1.721 14.440 84.770 -> 11.398 -> 1094.162 MByte/s p29 method 2 =non-blk :( 51.449) 0.019 0.303 4.624 28.121 52.878 84.770 -> 25.821 -> 2478.845 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 32.625) 0.031 0.475 6.782 34.578 53.395 82.639 -> 28.141 -> 2701.531 MByte/s p30 method 1 =Alltoal :(1610.303) 0.001 0.010 0.158 1.763 13.978 82.639 -> 11.154 -> 1070.783 MByte/s p30 method 2 =non-blk :( 50.899) 0.020 0.306 4.642 31.787 43.178 82.639 -> 24.715 -> 2372.661 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 32.575) 0.031 0.473 6.769 32.719 51.441 83.320 -> 27.596 -> 2649.222 MByte/s p31 method 1 =Alltoal :(1610.804) 0.001 0.010 0.160 1.738 14.038 83.320 -> 11.128 -> 1068.261 MByte/s p31 method 2 =non-blk :( 50.837) 0.020 0.304 4.557 30.713 33.678 83.320 -> 23.686 -> 2273.829 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 32.688) 0.031 0.476 6.851 34.420 54.956 79.862 -> 28.155 -> 2702.925 MByte/s p32 method 1 =Alltoal :(1614.404) 0.001 0.010 0.158 1.771 14.960 79.862 -> 11.153 -> 1070.677 MByte/s p32 method 2 =non-blk :( 50.878) 0.020 0.303 4.655 31.525 38.055 79.862 -> 24.226 -> 2325.686 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 32.637) 0.031 0.476 6.777 34.958 57.087 83.606 -> 28.326 -> 2719.307 MByte/s p33 method 1 =Alltoal :(1612.496) 0.001 0.010 0.158 1.760 14.565 83.606 -> 11.171 -> 1072.397 MByte/s p33 method 2 =non-blk :( 50.704) 0.020 0.306 4.612 31.534 47.292 83.606 -> 25.933 -> 2489.615 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 32.806) 0.030 0.476 6.840 34.215 49.280 82.578 -> 27.548 -> 2644.563 MByte/s p34 method 1 =Alltoal :(1611.900) 0.001 0.010 0.159 1.742 14.339 82.578 -> 11.000 -> 1056.033 MByte/s p34 method 2 =non-blk :( 51.020) 0.020 0.303 4.619 31.482 38.437 82.578 -> 23.630 -> 2268.506 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 32.938) 0.030 0.472 6.753 33.700 53.227 81.278 -> 27.483 -> 2638.345 MByte/s p35 method 1 =Alltoal :(1612.699) 0.001 0.010 0.157 1.706 12.447 81.278 -> 10.712 -> 1028.313 MByte/s p35 method 2 =non-blk :( 50.979) 0.020 0.302 4.624 30.632 49.162 81.278 -> 25.141 -> 2413.497 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 32.710) 0.031 0.471 6.907 41.517 73.997 105.684 -> 34.816 -> 3342.349 MByte/s p36 method 1 =Alltoal :(1630.402) 0.001 0.010 0.159 1.820 15.920 105.684 -> 12.685 -> 1217.784 MByte/s p36 method 2 =non-blk :( 51.849) 0.019 0.297 4.589 35.322 54.256 105.684 -> 28.652 -> 2750.587 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 11.475) 0.044 0.667 9.554 80.678 386.899 631.864 -> 165.703 -> 15907.471 MByte/s p37 method 1 =Alltoal :(1637.399) 0.000 0.005 0.078 1.249 18.527 631.864 -> 63.349 -> 6081.525 MByte/s p37 method 2 =non-blk :( 17.969) 0.028 0.453 6.628 72.451 373.695 631.864 -> 165.959 -> 15932.027 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 27.000) 0.019 0.288 4.089 33.139 41.183 114.453 -> 27.265 -> 2617.421 MByte/s p38 method 1 =Alltoal :(1630.604) 0.000 0.005 0.078 1.221 28.904 114.453 -> 19.356 -> 1858.220 MByte/s p38 method 2 =non-blk :( 26.378) 0.019 0.277 4.465 25.629 54.908 114.453 -> 31.087 -> 2984.359 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 11.113) 0.001 0.014 0.204 2.014 8.837 13.348 -> 3.759 -> 360.903 MByte/s p39 method 1 =Alltoal :(-------) 0.000 0.000 0.000 0.000 0.000 13.348 -> 1.256 -> 120.529 MByte/s p39 method 2 =non-blk :(-------) 0.000 0.000 0.000 0.000 0.000 13.348 -> 1.256 -> 120.529 MByte/s p40 acyclic-2dim-all p40 method 0 =Sndrcv :( 24.109) 0.037 0.581 8.353 54.697 95.999 159.554 -> 47.659 -> 4575.282 MByte/s p40 method 1 =Alltoal :(819.552) 0.001 0.018 0.279 2.866 31.513 159.554 -> 21.736 -> 2086.684 MByte/s p40 method 2 =non-blk :( 43.868) 0.020 0.324 4.905 43.351 84.906 159.554 -> 43.214 -> 4148.523 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 24.783) 0.031 0.488 6.992 49.123 119.729 153.160 -> 48.128 -> 4620.308 MByte/s p41 method 1 =Alltoal :(547.198) 0.001 0.023 0.365 3.784 38.995 153.160 -> 23.189 -> 2226.149 MByte/s p41 method 2 =non-blk :( 42.792) 0.018 0.287 4.352 39.787 102.238 153.160 -> 44.671 -> 4288.400 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 32.919) 0.030 0.470 6.944 39.143 54.200 106.791 -> 31.363 -> 3010.846 MByte/s p42 method 1 =Alltoal :(1636.100) 0.001 0.010 0.155 1.574 16.487 106.791 -> 13.555 -> 1301.277 MByte/s p42 method 2 =non-blk :( 51.204) 0.020 0.304 4.549 33.510 49.503 106.791 -> 27.558 -> 2645.609 MByte/s p43 cyclic-2dim-y p43 method 0 =Sndrcv :( 15.413) 0.065 1.021 14.374 108.650 365.612 623.364 -> 171.155 -> 16430.907 MByte/s p43 method 1 =Alltoal :(1632.392) 0.001 0.010 0.158 2.503 22.361 623.364 -> 62.913 -> 6039.626 MByte/s p43 method 2 =non-blk :( 36.470) 0.027 0.450 6.570 74.983 357.500 623.364 -> 156.984 -> 15070.441 MByte/s p44 cyclic-2dim-all p44 method 0 =Sndrcv :( 24.428) 0.041 0.641 9.131 59.628 103.425 156.884 -> 52.101 -> 5001.687 MByte/s p44 method 1 =Alltoal :(819.451) 0.001 0.020 0.303 3.157 29.635 156.884 -> 21.786 -> 2091.492 MByte/s p44 method 2 =non-blk :( 44.290) 0.023 0.358 5.419 47.127 88.469 156.884 -> 45.344 -> 4353.032 MByte/s p45 cyclic-3dim-x p45 method 0 =Sndrcv :( 32.687) 0.031 0.475 7.018 39.978 51.734 105.360 -> 31.872 -> 3059.699 MByte/s p45 method 1 =Alltoal :(1638.496) 0.001 0.010 0.152 1.567 15.832 105.360 -> 13.525 -> 1298.418 MByte/s p45 method 2 =non-blk :( 51.709) 0.019 0.299 4.638 33.185 42.147 105.360 -> 26.445 -> 2538.739 MByte/s p46 cyclic-3dim-y p46 method 0 =Sndrcv :( 27.451) 0.036 0.567 7.981 61.199 154.921 223.923 -> 67.103 -> 6441.847 MByte/s p46 method 1 =Alltoal :(1641.893) 0.001 0.010 0.157 2.185 22.179 223.923 -> 26.406 -> 2534.955 MByte/s p46 method 2 =non-blk :( 45.520) 0.022 0.351 5.207 47.647 143.857 223.923 -> 63.975 -> 6141.625 MByte/s p47 cyclic-3dim-z p47 method 0 =Sndrcv :( 15.381) 0.065 1.021 14.460 108.317 384.554 635.524 -> 175.182 -> 16817.489 MByte/s p47 method 1 =Alltoal :(1639.700) 0.001 0.010 0.156 2.501 25.536 635.524 -> 64.559 -> 6197.707 MByte/s p47 method 2 =non-blk :( 37.132) 0.027 0.434 6.514 74.472 370.357 635.524 -> 162.814 -> 15630.184 MByte/s p48 cyclic-3dim-all p48 method 0 =Sndrcv :( 25.417) 0.039 0.617 8.700 59.589 141.950 194.713 -> 58.472 -> 5613.299 MByte/s p48 method 1 =Alltoal :(543.932) 0.002 0.030 0.454 4.605 42.613 194.713 -> 28.245 -> 2711.553 MByte/s p48 method 2 =non-blk :( 44.340) 0.023 0.361 5.432 46.888 87.518 194.713 -> 49.007 -> 4704.690 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.049 0.769 10.721 83.060 313.114 530.401 || 140.428 -> 13481.074 MByte/s - ring, method 1 = Alltoal: 0.001 0.009 0.139 2.196 21.133 530.401 || 52.331 -> 5023.732 MByte/s - ring, method 2 = non-blk: 0.025 0.397 5.971 60.523 293.739 530.401 || 131.030 -> 12578.912 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.031 0.475 6.819 34.628 54.763 81.532 || 27.930 -> 2681.245 MByte/s - random, method 1 = Alltoal: 0.001 0.010 0.159 1.759 14.211 81.532 || 10.954 -> 1051.621 MByte/s - random, method 2 = non-blk: 0.020 0.304 4.619 31.084 45.616 81.532 || 24.724 -> 2373.543 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.039 0.604 8.550 53.630 130.947 207.954 || 62.627 -> 6012.159 MByte/s - average, method 1 = Alltoal: 0.001 0.009 0.149 1.965 17.330 207.954 || 23.943 -> 2298.491 MByte/s - average, method 2 = non-blk: 0.022 0.347 5.252 43.374 115.755 207.954 || 56.918 -> 5464.118 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 3.728 57.980 820.787 5148.524 12570.945 19963.584 || 6012.159 MByte/s - accumulated, mthd 1 = Alltoal: 0.056 0.897 14.267 188.686 1663.664 19963.584 || 2298.491 MByte/s - accumulated, mthd 2 = non-blk: 2.116 33.349 504.178 4163.907 11112.478 19963.584 || 5464.118 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: | effective| average rings random method_0 method_1 method_2 | bandwidth| crt,rnd only only Sndrcv Alltoal non-blk 1 3.728 0.039 0.049 0.031 0.039 0.001 0.022 2 7.417 0.077 0.098 0.061 0.077 0.001 0.044 4 14.256 0.149 0.190 0.116 0.149 0.002 0.086 8 29.255 0.305 0.389 0.239 0.305 0.005 0.175 16 57.980 0.604 0.769 0.475 0.604 0.009 0.347 32 114.881 1.197 1.519 0.943 1.197 0.019 0.693 64 226.468 2.359 2.979 1.868 2.359 0.037 1.371 128 418.225 4.357 5.488 3.458 4.357 0.074 2.654 256 820.787 8.550 10.721 6.819 8.550 0.149 5.252 512 1607.114 16.741 20.832 13.453 16.741 0.296 10.381 1024 3101.307 32.305 39.505 26.418 32.305 0.591 20.431 2048 3437.868 35.811 51.015 25.138 35.811 1.049 27.765 4096 5148.524 53.630 83.060 34.628 53.630 1.965 43.374 10624 6179.402 64.369 125.332 33.059 61.123 3.495 58.029 27554 9256.002 96.417 200.966 46.257 94.985 7.184 90.061 71468 10158.745 105.820 252.673 44.318 97.936 10.315 100.551 185364 12639.132 131.658 314.919 55.042 130.947 17.330 115.755 480774 14906.069 155.272 402.187 59.945 155.272 21.345 128.546 1246974 18461.402 192.306 484.900 76.267 192.225 23.166 160.228 3234251 18164.082 189.209 456.436 78.434 189.209 189.209 189.209 8388608 19963.584 207.954 530.401 81.532 207.954 207.954 207.954 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 185364 8388608 -> average -> accumulated (latency of one sendrecv (or equivalent method) in microsec) p00 ring-48*2fix :( 15.347) 0.065 1.012 13.424 106.607 386.130 635.981 -> 177.650 -> 17054.398 MByte/s p01 ring-24*4fix :( 15.537) 0.064 1.018 14.518 108.003 382.615 629.563 -> 174.216 -> 16724.709 MByte/s p02 ring-12*8fix :( 15.438) 0.065 1.024 14.489 108.278 371.109 634.396 -> 170.157 -> 16335.106 MByte/s p03 ring-4*24fix :( 26.544) 0.038 0.583 8.111 63.900 236.584 451.827 -> 114.474 -> 10989.480 MByte/s p04 ring-2*48fix :( 26.725) 0.037 0.583 8.254 64.274 261.270 425.375 -> 114.868 -> 11027.324 MByte/s p05 ring-1*96fix :( 26.806) 0.037 0.576 8.032 64.128 287.820 456.077 -> 119.947 -> 11514.926 MByte/s p06 random-cyc-1dim :( 32.869) 0.030 0.477 6.906 34.851 57.097 84.069 -> 28.753 -> 2760.324 MByte/s p07 random-cyc-1dim :( 33.123) 0.030 0.476 6.757 33.795 55.498 80.849 -> 27.610 -> 2650.516 MByte/s p08 random-cyc-1dim :( 32.663) 0.031 0.475 6.845 33.720 47.154 69.754 -> 26.389 -> 2533.365 MByte/s p09 random-cyc-1dim :( 32.594) 0.031 0.471 6.678 33.716 54.097 79.607 -> 27.041 -> 2595.924 MByte/s p10 random-cyc-1dim :( 32.612) 0.031 0.473 6.806 34.970 55.045 85.325 -> 28.725 -> 2757.594 MByte/s p11 random-cyc-1dim :( 32.794) 0.030 0.475 6.856 34.166 53.151 78.694 -> 27.608 -> 2650.323 MByte/s p12 random-cyc-1dim :( 32.656) 0.031 0.477 6.731 34.076 54.359 81.316 -> 27.262 -> 2617.111 MByte/s p13 random-cyc-1dim :( 32.600) 0.031 0.471 6.764 34.491 54.606 82.141 -> 27.688 -> 2658.081 MByte/s p14 random-cyc-1dim :( 32.350) 0.031 0.476 6.988 39.004 60.846 91.550 -> 30.679 -> 2945.168 MByte/s p15 random-cyc-1dim :( 32.543) 0.031 0.477 6.792 36.806 62.151 87.939 -> 30.386 -> 2917.090 MByte/s p16 random-cyc-1dim :( 32.638) 0.031 0.477 6.885 35.818 55.891 84.847 -> 28.583 -> 2743.989 MByte/s p17 random-cyc-1dim :( 32.593) 0.031 0.473 6.827 34.313 55.054 71.058 -> 26.799 -> 2572.741 MByte/s p18 random-cyc-1dim :( 32.593) 0.031 0.471 6.838 34.170 59.081 80.798 -> 28.138 -> 2701.243 MByte/s p19 random-cyc-1dim :( 32.562) 0.031 0.476 6.823 35.328 55.993 82.750 -> 28.497 -> 2735.687 MByte/s p20 random-cyc-1dim :( 32.694) 0.031 0.475 6.622 31.875 49.808 79.735 -> 26.203 -> 2515.484 MByte/s p21 random-cyc-1dim :( 32.746) 0.031 0.474 6.904 34.154 53.565 83.089 -> 27.775 -> 2666.440 MByte/s p22 random-cyc-1dim :( 32.593) 0.031 0.472 6.851 34.293 59.459 74.402 -> 27.168 -> 2608.140 MByte/s p23 random-cyc-1dim :( 32.549) 0.031 0.474 6.910 35.507 55.102 73.939 -> 26.953 -> 2587.481 MByte/s p24 random-cyc-1dim :( 32.649) 0.031 0.477 6.748 34.060 57.701 79.343 -> 27.865 -> 2675.031 MByte/s p25 random-cyc-1dim :( 33.049) 0.030 0.475 6.751 33.823 57.908 82.913 -> 28.377 -> 2724.212 MByte/s p26 random-cyc-1dim :( 32.613) 0.031 0.471 6.980 35.206 58.318 84.835 -> 29.066 -> 2790.379 MByte/s p27 random-cyc-1dim :( 32.619) 0.031 0.472 6.781 34.611 52.430 83.641 -> 27.833 -> 2671.932 MByte/s p28 random-cyc-1dim :( 32.607) 0.031 0.478 6.935 37.188 57.647 89.477 -> 29.889 -> 2869.322 MByte/s p29 random-cyc-1dim :( 32.581) 0.031 0.477 6.821 35.034 52.878 84.770 -> 28.641 -> 2749.515 MByte/s p30 random-cyc-1dim :( 32.625) 0.031 0.475 6.782 34.578 53.395 82.639 -> 28.263 -> 2713.272 MByte/s p31 random-cyc-1dim :( 32.575) 0.031 0.473 6.769 32.719 51.441 83.320 -> 27.708 -> 2659.977 MByte/s p32 random-cyc-1dim :( 32.688) 0.031 0.476 6.851 34.420 54.956 79.862 -> 28.232 -> 2710.245 MByte/s p33 random-cyc-1dim :( 32.637) 0.031 0.476 6.777 34.958 57.087 83.606 -> 28.650 -> 2750.426 MByte/s p34 random-cyc-1dim :( 32.806) 0.030 0.476 6.840 34.215 49.280 82.578 -> 27.692 -> 2658.427 MByte/s p35 random-cyc-1dim :( 32.938) 0.030 0.472 6.753 33.700 53.227 81.278 -> 27.525 -> 2642.424 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 32.710) 0.031 0.471 6.907 41.517 73.997 105.684 -> 34.833 -> 3343.957 MByte/s p37 best bi-section :( 11.475) 0.044 0.667 9.554 80.678 386.899 631.864 -> 168.937 -> 16217.913 MByte/s p38 worst bi-section :( 26.378) 0.019 0.288 4.465 33.139 54.908 114.453 -> 32.666 -> 3135.972 MByte/s p39 one PingPong Pair :( 11.113) 0.001 0.014 0.204 2.014 8.837 13.348 -> 3.759 -> 360.903 MByte/s p40 acyclic-2dim-all :( 24.109) 0.037 0.581 8.353 54.697 95.999 159.554 -> 48.654 -> 4670.760 MByte/s p41 acyclic-3dim-all :( 24.783) 0.031 0.488 6.992 49.123 119.729 153.160 -> 49.501 -> 4752.060 MByte/s p42 cyclic-2dim-x :( 32.919) 0.030 0.470 6.944 39.143 54.200 106.791 -> 31.527 -> 3026.635 MByte/s p43 cyclic-2dim-y :( 15.413) 0.065 1.021 14.374 108.650 365.612 623.364 -> 171.155 -> 16430.907 MByte/s p44 cyclic-2dim-all :( 24.428) 0.041 0.641 9.131 59.628 103.425 156.884 -> 53.083 -> 5095.953 MByte/s p45 cyclic-3dim-x :( 32.687) 0.031 0.475 7.018 39.978 51.734 105.360 -> 31.872 -> 3059.699 MByte/s p46 cyclic-3dim-y :( 27.451) 0.036 0.567 7.981 61.199 154.921 223.923 -> 69.216 -> 6644.704 MByte/s p47 cyclic-3dim-z :( 15.381) 0.065 1.021 14.460 108.317 384.554 635.524 -> 175.182 -> 16817.489 MByte/s p48 cyclic-3dim-all :( 25.417) 0.039 0.617 8.700 59.589 141.950 194.713 -> 58.472 -> 5613.299 MByte/s log_avg of all rings : 0.049 0.769 10.721 83.060 314.919 530.401 || 142.309 -> 13661.684 MByte/s log_avg of all random : 0.031 0.475 6.819 34.628 55.042 81.532 || 28.048 -> 2692.616 MByte/s log_avg(ring,random) : 0.039 0.604 8.550 53.630 131.658 207.954 || 63.178 -> 6065.118 MByte/s * size -> accumulated on all pr.: 3.728 57.980 820.787 5148.524 12639.132 19963.584 || 6065.118 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 6065.118 MByte/s on 96 processes ( = 63.178 MByte/s * 96 processes) Latency: 25.754 microsec Lmax: 8 MB b_eff at Lmax: 19963.584 MByte/s on 96 processes ( : 207.954 MByte/s * 96 processes) b_eff at Lmax (ring pattern): 50918.492 MByte/s on 96 processes ( : 530.401 MByte/s * 96 processes) Latency ring pattern: 0.211 microsec Ping-pong latency: 11.113 microsec Ping-pong bandwidth at Lmax: 1281.377 MByte/s at Lmax= 8.0 MB (MByte/s=1e6 Byte/s) (MB=2**20 Byte) system parameters : 96 nodes, 1024 MB/node system name : HI-UX/MPP hostname : hwwsr8k OS release : 03-07 OS version : 0 machine : SR8000 Date of measurement: Tue May 21 14:00:11 2002 Total execution wall clock time = 162 seconds | number | b_eff | Lmax | b_eff | b_eff | Latency | Latency | Latency | ping-pong | of pro | | | at Lmax | at Lmax | rings & | rings | ping- | bandwith | cessors | | | rings & | rings | random | only | pong | | | | | random | only | micro- | micro- | micro- | | | MByte/s | | MByte/s | MByte/s | sec | sec | sec | MByte/s -------------------------------------------------------------------------------------------------------------- | accumulated | 96 6065 8 MB 19964 50918 25.754 20.301 11.113 1281 | per process | 63 208 530 SECTION-BEFF-END b_eff = 6065.118 MB/s = 63.178 * 96 PEs with 1024 MB/PE on HI-UX/MPP hwwsr8k 03-07 0 SR8000