b_eff = 618.331 MB/s = 103.055 * 6 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000 SECTION-HEAD-BEGIN b_eff.c, Revision 3.3 from Nov. 29, 1999 MEMORY_PER_PROCESSOR = 1024 MBytes [1M = 1024*1024] 1-dim-paterns: size = 6 2-dim-paterns: size = 3 * 2 3-dim-paterns: size = 3 * 2 * 1 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-3*2fix 1=ring-1*6fix 2=ring-1*6fix 3=ring-1*6fix 4=ring-1*6fix 5=ring-1*6fix 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-all SECTION-HEAD-END SECTION-LOOP-BEGIN measurment loop: 75.441 sec SECTION-LOOP-END SECTION-LOOPLNGS-BEGIN method=0 | ------- repetition 0 --------- ------- repetition 1 --------- ------- repetition 2 --------- [sec] (= Sndrcv)| loop total per pattrn&mthd loop total per pattrn&mthd loop total per pattrn&mthd msg length| lng time minimum maximum lng time minimum maximum lng time minimum maximum ----------| ---- ------- ------- --------- ---- ------- ------- --------- ---- ------- ------- --------- 1 300 8.0e-01 8.2e-03 3.0e-02 136 3.6e-01 3.7e-03 1.4e-02 136 3.6e-01 3.7e-03 1.4e-02 2 150 4.0e-01 4.1e-03 1.5e-02 91 2.4e-01 2.5e-03 9.2e-03 91 2.4e-01 2.5e-03 9.3e-03 4 91 2.4e-01 2.5e-03 9.4e-03 91 2.4e-01 2.5e-03 9.3e-03 91 2.4e-01 2.5e-03 9.3e-03 8 90 2.4e-01 2.5e-03 9.2e-03 90 2.4e-01 2.5e-03 9.2e-03 90 2.4e-01 2.5e-03 9.2e-03 16 91 2.5e-01 2.6e-03 9.7e-03 91 2.5e-01 2.6e-03 9.7e-03 91 2.5e-01 2.5e-03 9.7e-03 32 87 2.4e-01 2.5e-03 9.5e-03 88 2.5e-01 2.5e-03 9.5e-03 89 2.5e-01 2.5e-03 9.6e-03 64 88 2.5e-01 2.5e-03 9.8e-03 88 2.5e-01 2.5e-03 9.8e-03 88 2.5e-01 2.5e-03 9.7e-03 128 86 2.6e-01 2.6e-03 1.0e-02 86 2.6e-01 2.6e-03 1.0e-02 86 2.6e-01 2.6e-03 1.0e-02 256 83 2.5e-01 2.5e-03 9.7e-03 83 2.6e-01 2.6e-03 9.9e-03 83 2.6e-01 2.6e-03 9.9e-03 512 82 2.5e-01 2.5e-03 9.8e-03 80 2.5e-01 2.5e-03 9.5e-03 80 2.5e-01 2.5e-03 9.7e-03 1024 80 2.5e-01 2.5e-03 9.8e-03 79 2.5e-01 2.5e-03 9.6e-03 80 2.5e-01 2.5e-03 9.7e-03 2048 79 4.0e-01 4.3e-03 1.5e-02 79 4.0e-01 4.1e-03 1.5e-02 79 4.0e-01 4.1e-03 1.5e-02 4096 45 2.8e-01 2.9e-03 1.1e-02 48 3.1e-01 3.3e-03 1.1e-02 48 3.0e-01 3.1e-03 1.1e-02 10624 29 3.3e-01 3.0e-03 1.1e-02 28 3.2e-01 2.8e-03 1.1e-02 29 3.4e-01 2.9e-03 1.1e-02 27554 18 3.6e-01 2.9e-03 1.2e-02 19 3.7e-01 3.1e-03 1.2e-02 19 3.8e-01 3.3e-03 1.2e-02 71468 11 4.1e-01 3.6e-03 1.3e-02 11 4.1e-01 3.7e-03 1.3e-02 11 4.1e-01 3.7e-03 1.3e-02 185364 5 3.6e-01 3.6e-03 1.0e-02 5 3.6e-01 3.1e-03 1.2e-02 5 3.6e-01 3.2e-03 1.2e-02 480774 2 3.1e-01 3.0e-03 1.0e-02 3 4.7e-01 4.5e-03 1.6e-02 3 4.7e-01 4.5e-03 1.7e-02 1246974 1 3.7e-01 3.5e-03 1.3e-02 1 3.6e-01 3.5e-03 1.2e-02 1 3.5e-01 3.5e-03 1.3e-02 3234251 1 8.8e-01 8.5e-03 3.0e-02 1 8.5e-01 8.5e-03 2.9e-02 1 5.7e-01 8.5e-03 3.2e-02 M 8388608 1 2.1e+00 2.1e-02 7.5e-02 1 2.1e+00 2.1e-02 7.3e-02 0 0.0e+00 1.0e+30 0.0e+00 method=1 | ------- repetition 0 --------- ------- repetition 1 --------- ------- repetition 2 --------- [sec] (=Alltoal)| loop total per pattrn&mthd loop total per pattrn&mthd loop total per pattrn&mthd msg length| lng time minimum maximum lng time minimum maximum lng time minimum maximum ----------| ---- ------- ------- --------- ---- ------- ------- --------- ---- ------- ------- --------- 1 300 2.1e+00 4.4e-02 4.7e-02 27 1.9e-01 3.9e-03 4.3e-03 25 1.8e-01 3.6e-03 4.0e-03 2 150 1.1e+00 2.2e-02 2.4e-02 17 1.2e-01 2.5e-03 2.8e-03 17 1.2e-01 2.5e-03 2.7e-03 4 75 5.3e-01 1.1e-02 1.2e-02 17 1.2e-01 2.5e-03 2.8e-03 17 1.2e-01 2.5e-03 2.8e-03 8 37 2.6e-01 5.4e-03 5.9e-03 17 1.2e-01 2.5e-03 2.7e-03 17 1.2e-01 2.5e-03 2.7e-03 16 18 1.3e-01 2.6e-03 3.0e-03 17 1.2e-01 2.5e-03 2.8e-03 17 1.2e-01 2.5e-03 2.8e-03 32 17 1.2e-01 2.5e-03 2.8e-03 17 1.2e-01 2.5e-03 2.8e-03 17 1.2e-01 2.5e-03 2.8e-03 64 16 1.2e-01 2.4e-03 2.7e-03 16 1.2e-01 2.4e-03 2.6e-03 16 1.2e-01 2.4e-03 2.7e-03 128 16 1.2e-01 2.4e-03 2.6e-03 16 1.2e-01 2.4e-03 2.6e-03 16 1.2e-01 2.4e-03 2.6e-03 256 16 1.2e-01 2.4e-03 2.8e-03 16 1.2e-01 2.4e-03 2.8e-03 16 1.2e-01 2.4e-03 2.7e-03 512 16 1.2e-01 2.4e-03 2.8e-03 16 1.2e-01 2.4e-03 2.7e-03 16 1.2e-01 2.5e-03 2.8e-03 1024 16 1.2e-01 2.5e-03 2.9e-03 16 1.2e-01 2.5e-03 2.8e-03 16 1.2e-01 2.5e-03 2.9e-03 2048 16 1.6e-01 2.8e-03 3.9e-03 16 1.6e-01 2.8e-03 4.1e-03 16 1.6e-01 2.8e-03 4.1e-03 4096 14 1.6e-01 2.6e-03 4.3e-03 14 1.6e-01 2.6e-03 4.4e-03 14 1.6e-01 2.6e-03 4.3e-03 10624 10 1.6e-01 2.3e-03 4.9e-03 10 1.6e-01 2.3e-03 4.8e-03 10 1.6e-01 2.3e-03 4.7e-03 27554 8 2.0e-01 2.4e-03 6.2e-03 8 2.0e-01 2.4e-03 6.1e-03 8 2.0e-01 2.4e-03 6.2e-03 71468 6 2.6e-01 2.7e-03 8.3e-03 6 2.6e-01 2.7e-03 8.4e-03 6 2.6e-01 2.7e-03 8.2e-03 185364 4 3.2e-01 2.8e-03 1.1e-02 4 3.2e-01 2.9e-03 1.1e-02 4 3.2e-01 2.9e-03 1.1e-02 480774 2 3.5e-01 3.3e-03 1.2e-02 2 3.5e-01 3.1e-03 1.2e-02 2 3.5e-01 3.1e-03 1.2e-02 1246974 1 4.0e-01 3.6e-03 1.4e-02 1 4.0e-01 3.7e-03 1.4e-02 1 4.0e-01 3.7e-03 1.4e-02 3234251 1 9.9e-01 9.1e-03 3.4e-02 1 9.7e-01 9.1e-03 3.3e-02 1 1.5e-01 9.9e-03 1.8e-02 M 8388608 1 2.4e+00 2.2e-02 8.3e-02 1 2.4e+00 2.2e-02 7.9e-02 0 0.0e+00 1.0e+30 0.0e+00 method=2 | ------- repetition 0 --------- ------- repetition 1 --------- ------- repetition 2 --------- [sec] (=non-blk)| loop total per pattrn&mthd loop total per pattrn&mthd loop total per pattrn&mthd msg length| lng time minimum maximum lng time minimum maximum lng time minimum maximum ----------| ---- ------- ------- --------- ---- ------- ------- --------- ---- ------- ------- --------- 1 300 1.3e+00 1.5e-02 4.2e-02 72 3.1e-01 3.6e-03 1.0e-02 73 3.2e-01 3.6e-03 1.0e-02 2 150 6.6e-01 7.8e-03 2.1e-02 49 2.1e-01 2.5e-03 6.8e-03 50 2.2e-01 2.5e-03 6.9e-03 4 75 3.3e-01 3.9e-03 1.1e-02 49 2.2e-01 2.5e-03 6.9e-03 49 2.2e-01 2.5e-03 6.8e-03 8 48 2.1e-01 2.5e-03 6.7e-03 49 2.1e-01 2.5e-03 6.8e-03 49 2.1e-01 2.5e-03 6.9e-03 16 48 2.2e-01 2.6e-03 7.1e-03 49 2.2e-01 2.6e-03 7.2e-03 49 2.2e-01 2.6e-03 7.1e-03 32 46 2.1e-01 2.5e-03 6.8e-03 47 2.2e-01 2.5e-03 6.9e-03 47 2.2e-01 2.5e-03 6.9e-03 64 46 2.2e-01 2.5e-03 6.9e-03 47 2.2e-01 2.5e-03 6.9e-03 46 2.1e-01 2.5e-03 6.8e-03 128 45 2.2e-01 2.5e-03 6.9e-03 46 2.2e-01 2.5e-03 7.0e-03 46 2.2e-01 2.5e-03 7.0e-03 256 44 2.1e-01 2.4e-03 6.9e-03 45 2.2e-01 2.4e-03 7.3e-03 45 2.2e-01 2.4e-03 7.3e-03 512 46 2.3e-01 2.5e-03 7.4e-03 46 2.3e-01 2.5e-03 7.4e-03 46 2.3e-01 2.5e-03 7.4e-03 1024 45 2.3e-01 2.5e-03 7.3e-03 46 2.3e-01 2.5e-03 7.4e-03 45 2.3e-01 2.5e-03 7.2e-03 2048 44 2.9e-01 3.2e-03 9.5e-03 45 2.9e-01 3.2e-03 9.6e-03 45 2.9e-01 3.2e-03 9.6e-03 4096 34 2.7e-01 2.9e-03 8.8e-03 35 2.8e-01 2.9e-03 9.3e-03 35 2.7e-01 2.9e-03 9.0e-03 10624 22 2.4e-01 2.6e-03 9.9e-03 22 2.4e-01 2.5e-03 7.8e-03 23 2.4e-01 2.7e-03 8.2e-03 27554 16 2.8e-01 2.9e-03 1.1e-02 17 2.9e-01 3.4e-03 9.7e-03 16 2.8e-01 3.1e-03 9.1e-03 71468 10 3.8e-01 3.2e-03 1.4e-02 9 3.4e-01 2.9e-03 1.2e-02 9 3.3e-01 2.8e-03 1.1e-02 185364 6 4.3e-01 3.6e-03 1.5e-02 6 4.2e-01 3.4e-03 1.5e-02 6 4.2e-01 3.5e-03 1.5e-02 480774 3 5.2e-01 4.3e-03 1.9e-02 3 5.1e-01 4.2e-03 2.0e-02 3 5.0e-01 4.0e-03 1.9e-02 1246974 1 3.9e-01 3.6e-03 1.5e-02 1 3.9e-01 3.6e-03 1.2e-02 1 4.0e-01 3.6e-03 1.4e-02 3234251 1 9.3e-01 8.9e-03 2.9e-02 1 9.3e-01 8.8e-03 3.2e-02 1 1.2e-01 9.2e-03 2.0e-02 M 8388608 1 2.3e+00 2.3e-02 7.8e-02 1 2.2e+00 2.0e-02 8.0e-02 0 0.0e+00 1.0e+30 0.0e+00 SECTION-LOOPLNGS-END SECTION-ELAPSED-BEGIN measurment loop: elapsed time = 75.441 sec sum of max elapsed time per entries above = 74.732 sec difference to elapsed time = 0.709 sec = 0.9% sum based on fastest repetition = 69.395 sec difference to elapsed time = 6.046 sec = 8.0% CAUTION: A difference above is more than 5 %. There may be problems with the MPI implementation or processes weren't on dedicated processors 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-3*2fix 1 6 1.00 1.00 0 ( -1 -1 1 ) p01 ring-1*6fix 2 12 2.00 1.00 0 ( -1 -1 0 ) p02 ring-1*6fix 2 12 2.00 1.00 0 ( -1 -1 0 ) p03 ring-1*6fix 2 12 2.00 1.00 0 ( -1 -1 0 ) p04 ring-1*6fix 2 12 2.00 1.00 0 ( -1 -1 0 ) p05 ring-1*6fix 2 12 2.00 1.00 0 ( -1 -1 0 ) p06 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p07 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 1 ) p08 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 2 ) p09 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p10 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 1 ) p11 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 1 ) p12 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 1 ) p13 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 1 ) p14 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p15 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 2 ) p16 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p17 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p18 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 2 ) p19 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 2 ) p20 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p21 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p22 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p23 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p24 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 1 ) p25 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 2 ) p26 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p27 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 1 ) p28 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 1 ) p29 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p30 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p31 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p32 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 1 ) p33 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p34 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 2 ) p35 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p36 worst-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 0 ) p37 best bi-section 2 6 1.00 0.50 0 ( -1 -1 0 ) p38 worst bi-section 2 6 1.00 0.50 0 ( -1 -1 0 ) p39 one PingPong Pair 2 2 1.00 0.50 4 ( -1 -1 0 ) p40 acyclic-2dim-all 4 14 2.33 0.58 0 ( -1 -1 0 ) p41 acyclic-3dim-all 4 14 2.33 0.58 0 ( -1 -1 0 ) p42 cyclic-2dim-x 2 12 2.00 1.00 0 ( -1 -1 0 ) p43 cyclic-2dim-y 1 6 1.00 1.00 0 ( -1 -1 2 ) p44 cyclic-2dim-all 3 18 3.00 1.00 0 ( -1 -1 0 ) p45 cyclic-3dim-x 2 12 2.00 1.00 0 ( -1 -1 0 ) p46 cyclic-3dim-y 1 6 1.00 1.00 0 ( -1 -1 2 ) p47 cyclic-3dim-all 3 18 3.00 1.00 0 ( -1 -1 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-3*2fix : 106.384 63.452 94.781 -> 106.384 -> 638.302 MByte/s p01 ring-1*6fix : 98.562 79.995 87.435 -> 98.562 -> 591.375 MByte/s p02 ring-1*6fix : 99.489 79.708 90.302 -> 99.489 -> 596.932 MByte/s p03 ring-1*6fix : 99.951 80.421 88.632 -> 99.951 -> 599.704 MByte/s p04 ring-1*6fix : 99.014 80.513 88.569 -> 99.014 -> 594.084 MByte/s p05 ring-1*6fix : 98.808 81.126 90.230 -> 98.808 -> 592.849 MByte/s p06 random-cyc-1dim : 99.053 80.246 89.664 -> 99.053 -> 594.320 MByte/s p07 random-cyc-1dim : 97.481 84.526 98.317 -> 98.317 -> 589.902 MByte/s p08 random-cyc-1dim : 134.051 102.623 129.083 -> 134.051 -> 804.303 MByte/s p09 random-cyc-1dim : 91.874 79.904 87.175 -> 91.874 -> 551.246 MByte/s p10 random-cyc-1dim : 93.089 76.721 90.639 -> 93.089 -> 558.533 MByte/s p11 random-cyc-1dim : 133.815 99.640 126.685 -> 133.815 -> 802.888 MByte/s p12 random-cyc-1dim : 98.676 76.328 88.718 -> 98.676 -> 592.057 MByte/s p13 random-cyc-1dim : 97.740 82.495 95.290 -> 97.740 -> 586.441 MByte/s p14 random-cyc-1dim : 92.620 79.952 91.601 -> 92.620 -> 555.717 MByte/s p15 random-cyc-1dim : 133.264 103.480 123.553 -> 133.264 -> 799.582 MByte/s p16 random-cyc-1dim : 98.531 85.460 96.817 -> 98.531 -> 591.185 MByte/s p17 random-cyc-1dim : 90.952 79.691 87.895 -> 90.952 -> 545.712 MByte/s p18 random-cyc-1dim : 101.357 85.795 95.661 -> 101.357 -> 608.142 MByte/s p19 random-cyc-1dim : 115.469 88.995 108.207 -> 115.469 -> 692.814 MByte/s p20 random-cyc-1dim : 98.653 80.359 90.517 -> 98.653 -> 591.919 MByte/s p21 random-cyc-1dim : 100.214 81.456 91.594 -> 100.214 -> 601.287 MByte/s p22 random-cyc-1dim : 92.037 80.049 88.410 -> 92.037 -> 552.223 MByte/s p23 random-cyc-1dim : 96.923 85.746 96.291 -> 96.923 -> 581.536 MByte/s p24 random-cyc-1dim : 96.279 84.614 97.785 -> 97.785 -> 586.710 MByte/s p25 random-cyc-1dim : 97.685 87.088 98.084 -> 98.084 -> 588.505 MByte/s p26 random-cyc-1dim : 98.134 84.841 96.040 -> 98.134 -> 588.803 MByte/s p27 random-cyc-1dim : 116.596 84.421 109.648 -> 116.596 -> 699.575 MByte/s p28 random-cyc-1dim : 96.849 82.690 98.705 -> 98.705 -> 592.231 MByte/s p29 random-cyc-1dim : 97.430 86.981 95.561 -> 97.430 -> 584.582 MByte/s p30 random-cyc-1dim : 93.550 79.164 89.658 -> 93.550 -> 561.298 MByte/s p31 random-cyc-1dim : 93.346 81.009 87.574 -> 93.346 -> 560.075 MByte/s p32 random-cyc-1dim : 90.842 77.108 88.657 -> 90.842 -> 545.054 MByte/s p33 random-cyc-1dim : 135.291 105.338 127.523 -> 135.291 -> 811.743 MByte/s p34 random-cyc-1dim : 113.259 89.366 109.542 -> 113.259 -> 679.554 MByte/s p35 random-cyc-1dim : 97.369 86.093 90.222 -> 97.369 -> 584.211 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 114.634 89.003 111.747 -> 114.634 -> 687.801 MByte/s p37 best bi-section : 99.692 70.381 99.927 -> 99.927 -> 599.562 MByte/s p38 worst bi-section : 100.227 93.970 112.703 -> 112.703 -> 676.219 MByte/s p39 one PingPong Pair : 34.186 6.049 6.049 -> 34.186 -> 205.119 MByte/s p40 acyclic-2dim-all : 86.929 75.187 83.263 -> 86.929 -> 521.576 MByte/s p41 acyclic-3dim-all : 86.256 75.637 83.333 -> 86.256 -> 517.539 MByte/s p42 cyclic-2dim-x : 99.650 76.022 88.510 -> 99.650 -> 597.901 MByte/s p43 cyclic-2dim-y : 106.024 69.482 94.342 -> 106.024 -> 636.145 MByte/s p44 cyclic-2dim-all : 100.687 82.232 91.651 -> 100.687 -> 604.121 MByte/s p45 cyclic-3dim-x : 102.294 76.683 89.108 -> 102.294 -> 613.764 MByte/s p46 cyclic-3dim-y : 104.863 69.464 94.078 -> 104.863 -> 629.178 MByte/s p47 cyclic-3dim-all : 99.658 82.701 89.904 -> 99.658 -> 597.948 MByte/s log_avg of all rings : 100.332 77.250 89.961 || 100.332 -> 601.992 MByte/s log_avg of all random : 102.269 85.088 98.122 || 102.430 -> 614.579 MByte/s log_avg(ring,random) : 101.296 81.075 93.953 ||(101.375 -> 608.253)MByte/s * size -> accumulated on all pr.: 607.775 486.448 563.719 ||(608.253)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-3*2fix : 99.059 105.027 98.436 -> 105.027 -> 630.161 MByte/s p01 ring-1*6fix : 88.087 97.782 97.781 -> 97.782 -> 586.690 MByte/s p02 ring-1*6fix : 92.411 99.617 97.810 -> 99.617 -> 597.699 MByte/s p03 ring-1*6fix : 96.123 98.117 100.376 -> 100.376 -> 602.259 MByte/s p04 ring-1*6fix : 98.850 97.419 98.251 -> 98.850 -> 593.099 MByte/s p05 ring-1*6fix : 97.590 97.028 98.528 -> 98.528 -> 591.166 MByte/s p06 random-cyc-1dim : 97.775 98.122 99.016 -> 99.016 -> 594.096 MByte/s p07 random-cyc-1dim : 90.347 97.168 100.428 -> 100.428 -> 602.569 MByte/s p08 random-cyc-1dim : 123.983 135.410 131.971 -> 135.410 -> 812.463 MByte/s p09 random-cyc-1dim : 90.793 90.487 93.046 -> 93.046 -> 558.277 MByte/s p10 random-cyc-1dim : 90.268 93.554 91.999 -> 93.554 -> 561.322 MByte/s p11 random-cyc-1dim : 131.200 134.421 125.814 -> 134.421 -> 806.523 MByte/s p12 random-cyc-1dim : 95.795 98.267 95.101 -> 98.267 -> 589.600 MByte/s p13 random-cyc-1dim : 98.093 101.018 96.742 -> 101.018 -> 606.110 MByte/s p14 random-cyc-1dim : 90.455 94.286 93.979 -> 94.286 -> 565.719 MByte/s p15 random-cyc-1dim : 131.166 132.950 131.331 -> 132.950 -> 797.702 MByte/s p16 random-cyc-1dim : 98.837 100.015 98.773 -> 100.015 -> 600.092 MByte/s p17 random-cyc-1dim : 93.423 91.422 91.456 -> 93.423 -> 560.541 MByte/s p18 random-cyc-1dim : 97.218 102.765 98.271 -> 102.765 -> 616.590 MByte/s p19 random-cyc-1dim : 112.357 113.863 115.652 -> 115.652 -> 693.912 MByte/s p20 random-cyc-1dim : 98.849 97.007 97.793 -> 98.849 -> 593.094 MByte/s p21 random-cyc-1dim : 95.422 98.669 100.520 -> 100.520 -> 603.120 MByte/s p22 random-cyc-1dim : 93.970 91.361 93.438 -> 93.970 -> 563.820 MByte/s p23 random-cyc-1dim : 98.417 99.403 99.609 -> 99.609 -> 597.652 MByte/s p24 random-cyc-1dim : 98.575 98.910 98.148 -> 98.910 -> 593.462 MByte/s p25 random-cyc-1dim : 99.176 99.039 101.194 -> 101.194 -> 607.164 MByte/s p26 random-cyc-1dim : 99.500 99.231 99.570 -> 99.570 -> 597.421 MByte/s p27 random-cyc-1dim : 115.618 113.164 109.834 -> 115.618 -> 693.711 MByte/s p28 random-cyc-1dim : 100.454 97.962 96.721 -> 100.454 -> 602.724 MByte/s p29 random-cyc-1dim : 98.754 97.862 99.794 -> 99.794 -> 598.761 MByte/s p30 random-cyc-1dim : 93.656 93.773 93.765 -> 93.773 -> 562.639 MByte/s p31 random-cyc-1dim : 93.604 94.287 95.336 -> 95.336 -> 572.017 MByte/s p32 random-cyc-1dim : 90.870 91.131 89.015 -> 91.131 -> 546.788 MByte/s p33 random-cyc-1dim : 135.600 134.053 133.314 -> 135.600 -> 813.597 MByte/s p34 random-cyc-1dim : 110.652 113.838 109.719 -> 113.838 -> 683.029 MByte/s p35 random-cyc-1dim : 99.120 97.708 99.121 -> 99.121 -> 594.728 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 114.753 114.582 115.173 -> 115.173 -> 691.040 MByte/s p37 best bi-section : 103.293 103.229 102.894 -> 103.293 -> 619.759 MByte/s p38 worst bi-section : 108.115 109.482 110.016 -> 110.016 -> 660.099 MByte/s p39 one PingPong Pair : 34.118 33.812 33.915 -> 34.118 -> 204.707 MByte/s p40 acyclic-2dim-all : 87.363 88.348 88.683 -> 88.683 -> 532.101 MByte/s p41 acyclic-3dim-all : 87.293 87.672 87.491 -> 87.672 -> 526.030 MByte/s p42 cyclic-2dim-x : 96.179 96.197 96.197 -> 96.197 -> 577.183 MByte/s p43 cyclic-2dim-y : 103.877 103.362 105.092 -> 105.092 -> 630.555 MByte/s p44 cyclic-2dim-all : 98.231 98.088 96.750 -> 98.231 -> 589.389 MByte/s p45 cyclic-3dim-x : 99.773 98.558 99.923 -> 99.923 -> 599.540 MByte/s p46 cyclic-3dim-y : 102.390 101.897 104.573 -> 104.573 -> 627.441 MByte/s p47 cyclic-3dim-all : 98.038 98.621 97.319 -> 98.621 -> 591.727 MByte/s log_avg of all rings : 95.270 99.128 98.526 || 100.002 -> 600.013 MByte/s log_avg of all random : 101.421 102.591 102.022 || 103.632 -> 621.789 MByte/s log_avg(ring,random) : 98.297 100.845 100.259 ||(101.801 -> 610.804)MByte/s * size -> accumulated on all pr.: 589.784 605.067 601.555 ||(610.804)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-3*2fix p00 method 0 =Sndrcv :( 27.323) 0.037 0.556 8.329 63.239 274.048 352.298 -> 106.384 -> 638.302 MByte/s p00 method 1 =Alltoal :(145.679) 0.007 0.108 1.674 20.777 174.297 223.643 -> 63.452 -> 380.711 MByte/s p00 method 2 =non-blk :( 50.263) 0.020 0.303 4.609 48.322 266.585 352.952 -> 94.781 -> 568.688 MByte/s p01 ring-1*6fix p01 method 0 =Sndrcv :( 27.552) 0.036 0.555 8.215 63.198 237.527 359.725 -> 98.562 -> 591.375 MByte/s p01 method 1 =Alltoal :( 72.945) 0.014 0.211 3.179 31.902 201.620 280.809 -> 79.995 -> 479.968 MByte/s p01 method 2 =non-blk :( 47.584) 0.021 0.320 4.714 47.408 203.903 316.199 -> 87.435 -> 524.608 MByte/s p02 ring-1*6fix p02 method 0 =Sndrcv :( 27.371) 0.037 0.563 8.308 63.459 225.587 372.744 -> 99.489 -> 596.932 MByte/s p02 method 1 =Alltoal :( 72.777) 0.014 0.211 3.091 31.241 201.044 280.331 -> 79.708 -> 478.247 MByte/s p02 method 2 =non-blk :( 47.247) 0.021 0.316 4.850 48.779 229.696 320.806 -> 90.302 -> 541.811 MByte/s p03 ring-1*6fix p03 method 0 =Sndrcv :( 27.522) 0.036 0.565 8.143 61.963 251.618 357.945 -> 99.951 -> 599.704 MByte/s p03 method 1 =Alltoal :( 73.166) 0.014 0.211 3.181 32.036 205.960 276.314 -> 80.421 -> 482.523 MByte/s p03 method 2 =non-blk :( 47.062) 0.021 0.323 4.845 47.415 217.711 328.392 -> 88.632 -> 531.791 MByte/s p04 ring-1*6fix p04 method 0 =Sndrcv :( 27.452) 0.036 0.556 8.174 62.604 234.253 368.034 -> 99.014 -> 594.084 MByte/s p04 method 1 =Alltoal :( 73.603) 0.014 0.212 3.194 31.499 204.908 277.282 -> 80.513 -> 483.077 MByte/s p04 method 2 =non-blk :( 47.171) 0.021 0.320 4.723 47.857 211.361 324.172 -> 88.569 -> 531.415 MByte/s p05 ring-1*6fix p05 method 0 =Sndrcv :( 27.618) 0.036 0.564 8.341 62.556 232.345 360.390 -> 98.808 -> 592.849 MByte/s p05 method 1 =Alltoal :( 73.480) 0.014 0.211 3.116 31.310 204.935 276.091 -> 81.126 -> 486.759 MByte/s p05 method 2 =non-blk :( 47.302) 0.021 0.319 4.835 48.096 230.291 322.775 -> 90.230 -> 541.378 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 27.500) 0.036 0.564 8.138 61.934 232.898 355.555 -> 99.053 -> 594.320 MByte/s p06 method 1 =Alltoal :( 73.061) 0.014 0.212 3.209 31.982 204.313 275.623 -> 80.246 -> 481.476 MByte/s p06 method 2 =non-blk :( 47.013) 0.021 0.323 4.855 47.127 222.127 323.423 -> 89.664 -> 537.985 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 27.529) 0.036 0.558 8.132 61.651 228.645 359.548 -> 97.481 -> 584.885 MByte/s p07 method 1 =Alltoal :( 77.579) 0.013 0.205 3.189 35.180 227.439 310.155 -> 84.526 -> 507.153 MByte/s p07 method 2 =non-blk :( 46.555) 0.021 0.325 4.789 48.431 256.765 354.122 -> 98.317 -> 589.902 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 24.337) 0.041 0.616 9.121 78.096 322.596 469.806 -> 134.051 -> 804.303 MByte/s p08 method 1 =Alltoal :( 78.521) 0.013 0.200 3.033 39.670 265.185 343.577 -> 102.623 -> 615.738 MByte/s p08 method 2 =non-blk :( 44.563) 0.022 0.345 5.181 55.566 314.041 483.704 -> 129.083 -> 774.498 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 27.485) 0.036 0.567 8.331 62.217 212.671 330.918 -> 91.874 -> 551.246 MByte/s p09 method 1 =Alltoal :( 72.740) 0.014 0.216 3.302 32.435 208.072 266.864 -> 79.904 -> 479.425 MByte/s p09 method 2 =non-blk :( 47.829) 0.021 0.323 4.851 48.139 204.408 311.879 -> 87.175 -> 523.048 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 27.294) 0.037 0.559 8.149 63.218 209.734 343.043 -> 93.089 -> 558.533 MByte/s p10 method 1 =Alltoal :( 72.537) 0.014 0.217 3.331 32.134 208.187 281.393 -> 76.721 -> 460.325 MByte/s p10 method 2 =non-blk :( 47.609) 0.021 0.320 4.786 49.290 249.537 317.096 -> 90.639 -> 543.833 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 24.407) 0.041 0.617 9.086 76.458 331.361 482.505 -> 133.815 -> 802.888 MByte/s p11 method 1 =Alltoal :( 78.426) 0.013 0.201 3.112 39.864 285.941 344.183 -> 99.640 -> 597.837 MByte/s p11 method 2 =non-blk :( 44.712) 0.022 0.344 5.185 55.042 317.720 439.230 -> 126.685 -> 760.109 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 27.393) 0.037 0.563 8.329 62.774 229.242 370.768 -> 98.676 -> 592.057 MByte/s p12 method 1 =Alltoal :( 73.340) 0.014 0.210 3.178 31.344 204.484 275.547 -> 76.328 -> 457.971 MByte/s p12 method 2 =non-blk :( 47.562) 0.021 0.323 4.852 47.955 228.302 323.348 -> 88.718 -> 532.305 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 27.309) 0.037 0.566 8.177 62.465 215.815 351.113 -> 97.740 -> 586.441 MByte/s p13 method 1 =Alltoal :( 77.541) 0.013 0.205 3.184 34.997 229.733 291.013 -> 82.495 -> 494.972 MByte/s p13 method 2 =non-blk :( 47.213) 0.021 0.326 4.846 48.003 255.913 334.047 -> 95.290 -> 571.740 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 27.512) 0.036 0.566 8.126 62.940 206.352 344.622 -> 92.620 -> 555.717 MByte/s p14 method 1 =Alltoal :( 72.341) 0.014 0.217 3.340 32.711 208.714 278.974 -> 79.952 -> 479.713 MByte/s p14 method 2 =non-blk :( 46.854) 0.021 0.318 4.790 48.048 230.864 332.044 -> 91.601 -> 549.609 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 24.180) 0.041 0.620 9.115 77.649 328.429 463.062 -> 133.264 -> 799.582 MByte/s p15 method 1 =Alltoal :( 78.980) 0.013 0.200 3.109 39.698 271.346 343.542 -> 103.480 -> 620.878 MByte/s p15 method 2 =non-blk :( 44.361) 0.023 0.348 5.210 56.598 320.145 444.136 -> 123.553 -> 741.319 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 27.320) 0.037 0.561 8.248 63.559 237.128 354.084 -> 98.531 -> 591.185 MByte/s p16 method 1 =Alltoal :( 77.241) 0.013 0.206 3.193 35.116 224.719 289.097 -> 85.460 -> 512.757 MByte/s p16 method 2 =non-blk :( 47.143) 0.021 0.325 4.899 47.849 253.603 329.372 -> 96.817 -> 580.904 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 27.511) 0.036 0.569 8.121 61.841 211.869 331.906 -> 90.952 -> 545.712 MByte/s p17 method 1 =Alltoal :( 72.444) 0.014 0.218 3.330 32.647 208.714 277.255 -> 79.691 -> 478.146 MByte/s p17 method 2 =non-blk :( 46.925) 0.021 0.317 4.785 47.517 228.327 309.852 -> 87.895 -> 527.372 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 27.487) 0.036 0.570 8.349 62.965 258.314 365.437 -> 101.357 -> 608.142 MByte/s p18 method 1 =Alltoal :( 77.647) 0.013 0.205 3.188 34.965 223.127 309.903 -> 85.795 -> 514.768 MByte/s p18 method 2 =non-blk :( 48.137) 0.021 0.324 4.796 48.391 238.081 335.430 -> 95.661 -> 573.966 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 27.342) 0.037 0.567 8.329 62.769 286.099 406.247 -> 115.469 -> 692.814 MByte/s p19 method 1 =Alltoal :( 73.722) 0.014 0.214 3.287 36.642 244.584 297.126 -> 88.995 -> 533.968 MByte/s p19 method 2 =non-blk :( 46.171) 0.022 0.334 4.981 49.595 279.616 385.674 -> 108.207 -> 649.242 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 27.416) 0.036 0.566 8.171 61.957 229.299 355.450 -> 98.653 -> 591.919 MByte/s p20 method 1 =Alltoal :( 73.055) 0.014 0.212 3.191 32.018 203.085 281.072 -> 80.359 -> 482.154 MByte/s p20 method 2 =non-blk :( 47.777) 0.021 0.318 4.767 46.694 234.093 331.861 -> 90.517 -> 543.101 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 27.243) 0.037 0.564 8.256 63.177 245.903 361.073 -> 100.214 -> 601.287 MByte/s p21 method 1 =Alltoal :( 73.221) 0.014 0.212 3.215 31.439 201.784 274.568 -> 81.456 -> 488.738 MByte/s p21 method 2 =non-blk :( 47.305) 0.021 0.323 4.725 48.555 214.209 327.975 -> 91.594 -> 549.564 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 27.280) 0.037 0.563 8.342 62.008 220.172 342.546 -> 92.037 -> 552.223 MByte/s p22 method 1 =Alltoal :( 72.238) 0.014 0.217 3.307 32.843 208.215 279.077 -> 80.049 -> 480.292 MByte/s p22 method 2 =non-blk :( 46.757) 0.021 0.324 4.836 48.364 219.388 297.976 -> 88.410 -> 530.459 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 27.415) 0.036 0.569 8.158 62.556 220.407 358.250 -> 96.923 -> 581.536 MByte/s p23 method 1 =Alltoal :( 78.157) 0.013 0.205 3.199 35.192 226.260 291.991 -> 85.746 -> 514.474 MByte/s p23 method 2 =non-blk :( 47.460) 0.021 0.324 4.894 48.523 244.514 342.967 -> 96.291 -> 577.747 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 27.382) 0.037 0.569 8.124 63.006 222.606 353.861 -> 96.279 -> 577.674 MByte/s p24 method 1 =Alltoal :( 78.080) 0.013 0.205 3.181 34.184 225.469 309.754 -> 84.614 -> 507.685 MByte/s p24 method 2 =non-blk :( 46.625) 0.021 0.328 4.874 49.358 245.650 326.812 -> 97.785 -> 586.710 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 27.550) 0.036 0.568 8.289 61.895 219.312 358.372 -> 97.685 -> 586.107 MByte/s p25 method 1 =Alltoal :( 78.005) 0.013 0.205 3.168 35.246 219.951 310.258 -> 87.088 -> 522.530 MByte/s p25 method 2 =non-blk :( 46.794) 0.021 0.328 4.803 47.723 248.338 342.106 -> 98.084 -> 588.505 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 27.492) 0.036 0.564 8.187 62.481 229.522 367.711 -> 98.134 -> 588.803 MByte/s p26 method 1 =Alltoal :( 77.075) 0.013 0.205 3.127 34.839 226.158 288.656 -> 84.841 -> 509.046 MByte/s p26 method 2 =non-blk :( 46.931) 0.021 0.323 4.910 48.473 246.631 335.155 -> 96.040 -> 576.242 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 27.303) 0.037 0.566 8.239 63.834 322.650 403.978 -> 116.596 -> 699.575 MByte/s p27 method 1 =Alltoal :( 73.540) 0.014 0.214 3.295 36.571 243.341 293.945 -> 84.421 -> 506.524 MByte/s p27 method 2 =non-blk :( 46.493) 0.022 0.329 4.878 49.223 274.342 407.917 -> 109.648 -> 657.888 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 27.533) 0.036 0.564 8.269 61.865 233.514 351.488 -> 96.849 -> 581.094 MByte/s p28 method 1 =Alltoal :( 77.206) 0.013 0.205 3.197 35.062 229.128 289.942 -> 82.690 -> 496.141 MByte/s p28 method 2 =non-blk :( 47.459) 0.021 0.322 4.801 47.676 238.562 357.190 -> 98.705 -> 592.231 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 27.441) 0.036 0.571 8.350 63.627 224.279 354.188 -> 97.430 -> 584.582 MByte/s p29 method 1 =Alltoal :( 77.389) 0.013 0.205 3.059 34.461 225.057 311.451 -> 86.981 -> 521.884 MByte/s p29 method 2 =non-blk :( 47.541) 0.021 0.324 4.895 48.984 252.136 349.343 -> 95.561 -> 573.364 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 27.357) 0.037 0.568 8.164 62.524 214.641 345.068 -> 93.550 -> 561.298 MByte/s p30 method 1 =Alltoal :( 72.298) 0.014 0.217 3.330 32.777 208.155 278.895 -> 79.164 -> 474.985 MByte/s p30 method 2 =non-blk :( 47.958) 0.021 0.319 4.795 48.067 220.432 314.640 -> 89.658 -> 537.948 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 27.373) 0.037 0.568 8.263 62.429 221.859 346.115 -> 93.346 -> 560.075 MByte/s p31 method 1 =Alltoal :( 72.420) 0.014 0.218 3.330 32.694 207.950 279.587 -> 81.009 -> 486.054 MByte/s p31 method 2 =non-blk :( 47.646) 0.021 0.323 4.706 48.390 223.578 311.763 -> 87.574 -> 525.446 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 27.302) 0.037 0.567 8.303 62.021 212.890 314.139 -> 90.842 -> 545.054 MByte/s p32 method 1 =Alltoal :( 72.923) 0.014 0.218 3.179 32.080 208.655 275.533 -> 77.108 -> 462.649 MByte/s p32 method 2 =non-blk :( 47.250) 0.021 0.319 4.831 47.306 221.882 318.190 -> 88.657 -> 531.943 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 24.295) 0.041 0.618 8.967 76.875 331.425 504.289 -> 135.291 -> 811.743 MByte/s p33 method 1 =Alltoal :( 78.978) 0.013 0.201 3.112 39.781 274.507 344.664 -> 105.338 -> 632.029 MByte/s p33 method 2 =non-blk :( 44.274) 0.023 0.344 5.237 54.833 322.888 440.785 -> 127.523 -> 765.135 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 27.340) 0.037 0.573 8.285 63.145 284.606 402.688 -> 113.259 -> 679.554 MByte/s p34 method 1 =Alltoal :( 73.521) 0.014 0.214 3.299 36.734 239.568 293.478 -> 89.366 -> 536.197 MByte/s p34 method 2 =non-blk :( 46.089) 0.022 0.331 4.917 48.953 285.871 396.072 -> 109.542 -> 657.254 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 27.529) 0.036 0.571 8.341 62.724 229.211 353.510 -> 97.369 -> 584.211 MByte/s p35 method 1 =Alltoal :( 77.300) 0.013 0.204 3.178 35.180 225.641 310.063 -> 86.093 -> 516.559 MByte/s p35 method 2 =non-blk :( 48.138) 0.021 0.328 4.846 47.494 247.042 327.820 -> 90.222 -> 541.329 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 27.242) 0.037 0.570 8.221 63.384 282.353 407.590 -> 114.634 -> 687.801 MByte/s p36 method 1 =Alltoal :( 73.574) 0.014 0.214 3.290 36.606 240.420 295.462 -> 89.003 -> 534.015 MByte/s p36 method 2 =non-blk :( 46.137) 0.022 0.329 4.992 49.614 289.593 396.559 -> 111.747 -> 670.479 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 23.838) 0.021 0.313 4.502 38.855 240.980 377.490 -> 99.692 -> 598.153 MByte/s p37 method 1 =Alltoal :( 72.641) 0.007 0.107 1.672 20.605 177.045 224.745 -> 70.381 -> 422.288 MByte/s p37 method 2 =non-blk :( 25.388) 0.020 0.300 4.698 47.676 243.258 351.547 -> 99.927 -> 599.562 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 23.935) 0.021 0.314 4.496 39.025 240.042 374.091 -> 100.227 -> 601.361 MByte/s p38 method 1 =Alltoal :( 72.961) 0.007 0.109 1.686 22.184 253.063 372.182 -> 93.970 -> 563.820 MByte/s p38 method 2 =non-blk :( 25.549) 0.020 0.301 4.741 48.303 318.310 412.928 -> 112.703 -> 676.219 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 23.158) 0.007 0.107 1.543 13.483 80.684 130.988 -> 34.186 -> 205.119 MByte/s p39 method 1 =Alltoal :(-------) 0.000 0.000 0.000 0.000 0.000 0.000 -> 6.049 -> 36.294 MByte/s p39 method 2 =non-blk :(-------) 0.000 0.000 0.000 0.000 0.000 0.000 -> 6.049 -> 36.294 MByte/s p40 acyclic-2dim-all p40 method 0 =Sndrcv :( 25.296) 0.023 0.351 5.110 40.358 223.063 304.626 -> 86.929 -> 521.576 MByte/s p40 method 1 =Alltoal :( 36.861) 0.016 0.242 3.641 33.900 202.464 256.950 -> 75.187 -> 451.125 MByte/s p40 method 2 =non-blk :( 32.247) 0.018 0.284 4.137 40.271 211.069 275.279 -> 83.263 -> 499.580 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 25.377) 0.023 0.350 5.100 40.440 226.212 306.572 -> 86.256 -> 517.539 MByte/s p41 method 1 =Alltoal :( 36.963) 0.016 0.242 3.651 33.900 202.205 256.623 -> 75.637 -> 453.821 MByte/s p41 method 2 =non-blk :( 32.120) 0.018 0.284 4.140 39.984 214.791 282.310 -> 83.333 -> 499.998 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 27.544) 0.036 0.570 8.350 62.935 232.839 333.443 -> 99.650 -> 597.901 MByte/s p42 method 1 =Alltoal :( 73.073) 0.014 0.211 3.204 30.922 207.981 251.438 -> 76.022 -> 456.135 MByte/s p42 method 2 =non-blk :( 47.306) 0.021 0.324 4.885 49.332 217.032 314.015 -> 88.510 -> 531.063 MByte/s p43 cyclic-2dim-y p43 method 0 =Sndrcv :( 27.265) 0.037 0.571 8.336 61.359 268.020 351.974 -> 106.024 -> 636.145 MByte/s p43 method 1 =Alltoal :(146.217) 0.007 0.108 1.682 20.769 176.789 221.406 -> 69.482 -> 416.894 MByte/s p43 method 2 =non-blk :( 50.136) 0.020 0.301 4.733 48.662 263.610 352.715 -> 94.342 -> 566.052 MByte/s p44 cyclic-2dim-all p44 method 0 =Sndrcv :( 27.460) 0.036 0.567 8.261 61.794 255.535 364.136 -> 100.687 -> 604.121 MByte/s p44 method 1 =Alltoal :( 49.437) 0.020 0.310 4.580 40.110 217.477 265.731 -> 82.232 -> 493.395 MByte/s p44 method 2 =non-blk :( 46.287) 0.022 0.331 4.887 47.920 226.298 316.842 -> 91.651 -> 549.905 MByte/s p45 cyclic-3dim-x p45 method 0 =Sndrcv :( 27.390) 0.037 0.569 8.239 62.935 226.745 368.560 -> 102.294 -> 613.764 MByte/s p45 method 1 =Alltoal :( 73.532) 0.014 0.212 3.186 30.519 204.427 247.248 -> 76.683 -> 460.100 MByte/s p45 method 2 =non-blk :( 46.746) 0.021 0.326 4.867 49.392 214.977 314.917 -> 89.108 -> 534.649 MByte/s p46 cyclic-3dim-y p46 method 0 =Sndrcv :( 27.309) 0.037 0.571 8.313 59.650 262.926 350.723 -> 104.863 -> 629.178 MByte/s p46 method 1 =Alltoal :(144.892) 0.007 0.108 1.677 20.716 180.184 222.971 -> 69.464 -> 416.783 MByte/s p46 method 2 =non-blk :( 51.346) 0.019 0.300 4.677 48.464 228.277 351.075 -> 94.078 -> 564.467 MByte/s p47 cyclic-3dim-all p47 method 0 =Sndrcv :( 27.464) 0.036 0.568 8.357 61.728 250.448 350.705 -> 99.658 -> 597.948 MByte/s p47 method 1 =Alltoal :( 49.079) 0.020 0.311 4.578 40.488 216.695 265.158 -> 82.701 -> 496.208 MByte/s p47 method 2 =non-blk :( 46.333) 0.022 0.330 4.885 48.254 233.145 330.451 -> 89.904 -> 539.423 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.036 0.560 8.251 62.834 242.048 361.794 || 100.332 -> 601.992 MByte/s - ring, method 1 = Alltoal: 0.012 0.189 2.836 29.464 198.463 268.227 || 77.250 -> 463.502 MByte/s - ring, method 2 = non-blk: 0.021 0.317 4.762 47.977 225.735 327.339 || 89.961 -> 539.766 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.037 0.573 8.344 64.379 242.755 370.799 || 102.269 -> 613.614 MByte/s - random, method 1 = Alltoal: 0.013 0.209 3.211 34.594 225.217 295.762 || 85.088 -> 510.531 MByte/s - random, method 2 = non-blk: 0.021 0.327 4.882 49.142 249.375 348.949 || 98.122 -> 588.735 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.037 0.566 8.297 63.602 242.401 366.269 || 101.296 -> 607.775 MByte/s - average, method 1 = Alltoal: 0.013 0.199 3.018 31.926 211.417 281.658 || 81.075 -> 486.448 MByte/s - average, method 2 = non-blk: 0.021 0.322 4.822 48.556 237.261 337.971 || 93.953 -> 563.719 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 0.220 3.397 49.784 381.611 1454.407 2197.614 || 607.775 MByte/s - accumulated, mthd 1 = Alltoal: 0.076 1.194 18.106 191.558 1268.503 1689.948 || 486.448 MByte/s - accumulated, mthd 2 = non-blk: 0.127 1.930 28.930 291.337 1423.565 2027.828 || 563.719 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 0.220 0.037 0.036 0.037 0.037 0.013 0.021 2 0.440 0.073 0.073 0.074 0.073 0.025 0.042 4 0.872 0.145 0.144 0.147 0.145 0.051 0.084 8 1.764 0.294 0.292 0.296 0.294 0.101 0.169 16 3.397 0.566 0.560 0.573 0.566 0.199 0.322 32 6.736 1.123 1.113 1.132 1.123 0.396 0.639 64 13.263 2.210 2.199 2.222 2.210 0.784 1.269 128 25.283 4.214 4.172 4.256 4.214 1.536 2.466 256 49.784 8.297 8.251 8.344 8.297 3.018 4.822 512 98.072 16.345 16.218 16.474 16.345 6.033 9.582 1024 190.884 31.814 31.611 32.018 31.814 11.894 18.851 2048 234.647 39.108 38.617 39.605 39.108 18.395 29.042 4096 381.611 63.602 62.834 64.379 63.602 31.926 48.556 10624 561.670 93.612 92.368 94.872 88.183 58.743 92.222 27554 886.073 147.679 144.118 151.328 131.355 100.110 143.688 71468 1093.910 182.318 177.633 187.127 179.394 148.610 170.538 185364 1491.657 248.610 242.778 254.582 242.401 211.417 237.261 480774 1690.918 281.820 274.976 288.834 279.496 255.320 246.130 1246974 1945.168 324.195 319.639 328.815 321.151 249.076 284.439 3234251 2100.391 350.065 345.642 354.545 349.377 320.654 341.585 8388608 2200.437 366.740 361.906 371.638 366.269 281.658 337.971 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-3*2fix :( 27.323) 0.037 0.556 8.329 63.239 274.048 352.952 -> 106.716 -> 640.298 MByte/s p01 ring-1*6fix :( 27.552) 0.036 0.555 8.215 63.198 237.527 359.725 -> 99.413 -> 596.479 MByte/s p02 ring-1*6fix :( 27.371) 0.037 0.563 8.308 63.459 229.696 372.744 -> 100.722 -> 604.329 MByte/s p03 ring-1*6fix :( 27.522) 0.036 0.565 8.143 61.963 251.618 357.945 -> 101.135 -> 606.808 MByte/s p04 ring-1*6fix :( 27.452) 0.036 0.556 8.174 62.604 234.253 368.034 -> 99.913 -> 599.476 MByte/s p05 ring-1*6fix :( 27.618) 0.036 0.564 8.341 62.556 232.345 360.390 -> 99.727 -> 598.360 MByte/s p06 random-cyc-1dim :( 27.500) 0.036 0.564 8.138 61.934 232.898 355.555 -> 100.181 -> 601.085 MByte/s p07 random-cyc-1dim :( 27.529) 0.036 0.558 8.132 61.651 256.765 359.548 -> 102.592 -> 615.554 MByte/s p08 random-cyc-1dim :( 24.337) 0.041 0.616 9.121 78.096 322.596 483.704 -> 136.394 -> 818.365 MByte/s p09 random-cyc-1dim :( 27.485) 0.036 0.567 8.331 62.217 212.671 330.918 -> 93.702 -> 562.213 MByte/s p10 random-cyc-1dim :( 27.294) 0.037 0.559 8.149 63.218 249.537 343.043 -> 96.301 -> 577.803 MByte/s p11 random-cyc-1dim :( 24.407) 0.041 0.617 9.086 76.458 331.361 482.505 -> 135.554 -> 813.326 MByte/s p12 random-cyc-1dim :( 27.393) 0.037 0.563 8.329 62.774 229.242 370.768 -> 100.495 -> 602.973 MByte/s p13 random-cyc-1dim :( 27.309) 0.037 0.566 8.177 62.465 255.913 351.113 -> 101.564 -> 609.383 MByte/s p14 random-cyc-1dim :( 27.512) 0.036 0.566 8.126 62.940 230.864 344.622 -> 95.448 -> 572.686 MByte/s p15 random-cyc-1dim :( 24.180) 0.041 0.620 9.115 77.649 328.429 463.062 -> 134.299 -> 805.793 MByte/s p16 random-cyc-1dim :( 27.320) 0.037 0.561 8.248 63.559 253.603 354.084 -> 101.248 -> 607.488 MByte/s p17 random-cyc-1dim :( 27.511) 0.036 0.569 8.121 61.841 228.327 331.906 -> 94.731 -> 568.389 MByte/s p18 random-cyc-1dim :( 27.487) 0.036 0.570 8.349 62.965 258.314 365.437 -> 103.331 -> 619.987 MByte/s p19 random-cyc-1dim :( 27.342) 0.037 0.567 8.329 62.769 286.099 406.247 -> 116.933 -> 701.600 MByte/s p20 random-cyc-1dim :( 27.416) 0.036 0.566 8.171 61.957 234.093 355.450 -> 99.838 -> 599.028 MByte/s p21 random-cyc-1dim :( 27.243) 0.037 0.564 8.256 63.177 245.903 361.073 -> 101.494 -> 608.964 MByte/s p22 random-cyc-1dim :( 27.280) 0.037 0.563 8.342 62.008 220.172 342.546 -> 94.527 -> 567.163 MByte/s p23 random-cyc-1dim :( 27.415) 0.036 0.569 8.158 62.556 244.514 358.250 -> 100.448 -> 602.686 MByte/s p24 random-cyc-1dim :( 27.382) 0.037 0.569 8.124 63.006 245.650 353.861 -> 101.872 -> 611.232 MByte/s p25 random-cyc-1dim :( 27.550) 0.036 0.568 8.289 61.895 248.338 358.372 -> 101.590 -> 609.541 MByte/s p26 random-cyc-1dim :( 27.492) 0.036 0.564 8.187 62.481 246.631 367.711 -> 101.430 -> 608.582 MByte/s p27 random-cyc-1dim :( 27.303) 0.037 0.566 8.239 63.834 322.650 407.917 -> 117.703 -> 706.219 MByte/s p28 random-cyc-1dim :( 27.533) 0.036 0.564 8.269 61.865 238.562 357.190 -> 101.127 -> 606.760 MByte/s p29 random-cyc-1dim :( 27.441) 0.036 0.571 8.350 63.627 252.136 354.188 -> 101.392 -> 608.350 MByte/s p30 random-cyc-1dim :( 27.357) 0.037 0.568 8.164 62.524 220.432 345.068 -> 95.182 -> 571.092 MByte/s p31 random-cyc-1dim :( 27.373) 0.037 0.568 8.263 62.429 223.578 346.115 -> 96.251 -> 577.506 MByte/s p32 random-cyc-1dim :( 27.302) 0.037 0.567 8.303 62.021 221.882 318.190 -> 92.649 -> 555.895 MByte/s p33 random-cyc-1dim :( 24.295) 0.041 0.618 8.967 76.875 331.425 504.289 -> 136.532 -> 819.193 MByte/s p34 random-cyc-1dim :( 27.340) 0.037 0.573 8.285 63.145 285.871 402.688 -> 114.494 -> 686.965 MByte/s p35 random-cyc-1dim :( 27.529) 0.036 0.571 8.341 62.724 247.042 353.510 -> 99.952 -> 599.710 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 27.242) 0.037 0.570 8.221 63.384 289.593 407.590 -> 118.348 -> 710.086 MByte/s p37 best bi-section :( 23.838) 0.021 0.313 4.698 47.676 243.258 377.490 -> 104.558 -> 627.350 MByte/s p38 worst bi-section :( 23.935) 0.021 0.314 4.741 48.303 318.310 412.928 -> 113.396 -> 680.376 MByte/s p39 one PingPong Pair :( 23.158) 0.007 0.107 1.543 13.483 80.684 130.988 -> 34.186 -> 205.119 MByte/s p40 acyclic-2dim-all :( 25.296) 0.023 0.351 5.110 40.358 223.063 304.626 -> 88.939 -> 533.633 MByte/s p41 acyclic-3dim-all :( 25.377) 0.023 0.350 5.100 40.440 226.212 306.572 -> 88.224 -> 529.346 MByte/s p42 cyclic-2dim-x :( 27.544) 0.036 0.570 8.350 62.935 232.839 333.443 -> 100.008 -> 600.047 MByte/s p43 cyclic-2dim-y :( 27.265) 0.037 0.571 8.336 61.359 268.020 352.715 -> 106.201 -> 637.206 MByte/s p44 cyclic-2dim-all :( 27.460) 0.036 0.567 8.261 61.794 255.535 364.136 -> 101.261 -> 607.565 MByte/s p45 cyclic-3dim-x :( 27.390) 0.037 0.569 8.239 62.935 226.745 368.560 -> 102.454 -> 614.722 MByte/s p46 cyclic-3dim-y :( 27.309) 0.037 0.571 8.313 59.650 262.926 351.075 -> 105.253 -> 631.518 MByte/s p47 cyclic-3dim-all :( 27.464) 0.036 0.568 8.357 61.728 250.448 350.705 -> 100.244 -> 601.462 MByte/s log_avg of all rings : 0.036 0.560 8.251 62.834 242.778 361.906 || 101.241 -> 607.444 MByte/s log_avg of all random : 0.037 0.573 8.344 64.379 254.582 371.638 || 104.902 -> 629.413 MByte/s log_avg(ring,random) : 0.037 0.566 8.297 63.602 248.610 366.740 || 103.055 -> 618.331 MByte/s * size -> accumulated on all pr.: 0.220 3.397 49.784 381.611 1491.657 2200.437 || 618.331 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 618.331 MByte/s on 6 processes ( = 103.055 MByte/s * 6 processes) Ping-pong latency: 23.158 microsec Ping-pong bandwidth: 785.927 MByte/s at Lmax= 8.000 MByte (MByte/s=1e6 Byte/s) (MByte=2**20 Byte) system parameters : 6 nodes, 1024 MB/node system name : HI-UX/MPP hostname : himiko OS release : 03-00 OS version : ad2b0 machine : SR8000 Date of measurement: Mon Nov 29 13:44:10 1999 Total execution wall clock time = 77 seconds SECTION-BEFF-END b_eff = 618.331 MB/s = 103.055 * 6 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000