b_eff = 768.112 MB/s = 48.007 * 16 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 = 16 2-dim-paterns: size = 4 * 4 3-dim-paterns: size = 4 * 2 * 2 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-8*2fix 1=ring-4*4fix 2=ring-2*8fix 3=ring-1*16fix 4=ring-1*16fix 5=ring-1*16fix 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 SECTION-HEAD-END SECTION-LOOP-BEGIN measurment loop: 136.752 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 9.1e-01 4.9e-03 3.4e-02 228 6.9e-01 3.8e-03 2.6e-02 226 6.9e-01 3.8e-03 2.6e-02 2 154 4.7e-01 2.5e-03 1.8e-02 150 4.6e-01 2.5e-03 1.7e-02 149 4.6e-01 2.5e-03 1.7e-02 4 152 4.7e-01 2.5e-03 1.8e-02 152 4.7e-01 2.5e-03 1.8e-02 151 4.7e-01 2.5e-03 1.8e-02 8 153 4.7e-01 2.5e-03 1.7e-02 151 4.6e-01 2.5e-03 1.7e-02 153 4.6e-01 2.5e-03 1.8e-02 16 152 5.0e-01 2.5e-03 1.8e-02 151 4.8e-01 2.5e-03 1.8e-02 152 4.8e-01 2.5e-03 1.8e-02 32 150 4.8e-01 2.6e-03 1.8e-02 150 4.8e-01 2.5e-03 1.8e-02 152 4.9e-01 2.6e-03 1.9e-02 64 146 4.8e-01 2.6e-03 1.8e-02 148 4.9e-01 2.6e-03 1.9e-02 148 4.9e-01 2.6e-03 1.9e-02 128 142 4.8e-01 2.7e-03 1.9e-02 143 4.9e-01 2.7e-03 1.9e-02 143 4.9e-01 2.8e-03 1.9e-02 256 130 4.4e-01 2.5e-03 1.7e-02 130 4.5e-01 2.5e-03 1.7e-02 129 4.5e-01 2.4e-03 1.7e-02 512 127 4.4e-01 2.6e-03 1.7e-02 130 4.6e-01 2.7e-03 1.7e-02 131 4.7e-01 2.6e-03 1.8e-02 1024 120 4.4e-01 2.5e-03 1.7e-02 121 4.4e-01 2.6e-03 1.7e-02 125 4.6e-01 2.7e-03 1.8e-02 2048 118 7.6e-01 3.2e-03 2.6e-02 117 7.4e-01 3.2e-03 2.6e-02 114 7.2e-01 3.1e-03 2.5e-02 4096 92 7.5e-01 3.3e-03 2.6e-02 91 7.9e-01 3.4e-03 2.6e-02 92 7.6e-01 3.3e-03 3.2e-02 10624 53 9.3e-01 3.0e-03 3.9e-02 51 9.0e-01 2.8e-03 3.8e-02 53 9.3e-01 3.0e-03 4.2e-02 27554 33 1.0e+00 3.0e-03 5.4e-02 35 1.1e+00 3.3e-03 5.6e-02 34 1.1e+00 2.9e-03 5.4e-02 71468 21 1.4e+00 4.2e-03 5.6e-02 20 1.4e+00 4.2e-03 5.8e-02 22 1.5e+00 4.7e-03 6.2e-02 185364 9 1.2e+00 4.2e-03 4.7e-02 9 1.2e+00 3.9e-03 5.5e-02 9 1.2e+00 3.4e-03 4.8e-02 480774 4 1.3e+00 4.9e-03 6.1e-02 4 1.3e+00 4.0e-03 6.1e-02 5 1.6e+00 4.6e-03 7.6e-02 1246974 1 7.9e-01 5.3e-03 4.1e-02 1 7.5e-01 2.9e-03 4.3e-02 2 1.5e+00 5.0e-03 8.1e-02 3234251 1 1.1e+00 1.4e-02 6.2e-02 M 1 1.0e+00 1.5e-02 6.4e-02 M 1 1.2e+00 1.4e-02 7.0e-02 M 8388608 1 2.5e+00 3.7e-02 1.5e-01 R 1 2.6e+00 3.5e-02 1.8e-01 R 0 0.0e+00 1.0e+30 0.0e+00 R Explanation: M = only the best method is used R = only best method and only 2 repetitions 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 6.0e+00 1.2e-01 1.4e-01 27 5.4e-01 1.1e-02 1.3e-02 9 1.8e-01 3.5e-03 4.3e-03 2 150 3.0e+00 5.9e-02 7.0e-02 13 2.6e-01 5.1e-03 6.0e-03 6 1.2e-01 2.3e-03 2.9e-03 4 75 1.5e+00 3.0e-02 3.5e-02 6 1.2e-01 2.3e-03 2.8e-03 6 1.2e-01 2.3e-03 2.8e-03 8 37 7.3e-01 1.5e-02 1.7e-02 6 1.2e-01 2.3e-03 2.8e-03 6 1.2e-01 2.3e-03 3.0e-03 16 18 3.6e-01 7.1e-03 8.4e-03 6 1.2e-01 2.3e-03 2.8e-03 6 1.2e-01 2.3e-03 2.8e-03 32 9 1.8e-01 3.5e-03 4.2e-03 6 1.2e-01 2.3e-03 2.8e-03 6 1.2e-01 2.3e-03 2.8e-03 64 6 1.2e-01 2.3e-03 2.8e-03 6 1.2e-01 2.3e-03 2.9e-03 6 1.2e-01 2.3e-03 3.0e-03 128 6 1.2e-01 2.4e-03 2.9e-03 6 1.2e-01 2.4e-03 3.0e-03 6 1.2e-01 2.4e-03 2.9e-03 256 6 1.2e-01 2.4e-03 2.8e-03 6 1.2e-01 2.4e-03 2.9e-03 6 1.2e-01 2.4e-03 2.8e-03 512 6 1.2e-01 2.4e-03 2.9e-03 6 1.2e-01 2.4e-03 2.8e-03 6 1.3e-01 2.4e-03 1.1e-02 1024 6 1.2e-01 2.4e-03 3.0e-03 6 1.2e-01 2.4e-03 2.9e-03 6 1.3e-01 2.4e-03 1.1e-02 2048 6 1.6e-01 2.6e-03 4.8e-03 6 1.5e-01 2.6e-03 3.6e-03 6 1.5e-01 2.5e-03 3.7e-03 4096 5 1.5e-01 2.2e-03 3.5e-03 5 1.5e-01 2.2e-03 3.9e-03 5 1.5e-01 2.3e-03 7.1e-03 10624 4 1.6e-01 2.1e-03 4.0e-03 4 1.6e-01 2.2e-03 6.2e-03 4 1.6e-01 2.1e-03 4.0e-03 27554 3 1.8e-01 1.9e-03 6.3e-03 3 1.8e-01 1.9e-03 5.2e-03 3 1.8e-01 1.9e-03 4.7e-03 71468 3 3.1e-01 3.1e-03 1.1e-02 3 3.2e-01 3.0e-03 1.8e-02 3 3.0e-01 3.0e-03 8.8e-03 185364 1 2.1e-01 1.6e-03 7.7e-03 1 2.0e-01 1.9e-03 7.0e-03 1 2.0e-01 1.9e-03 9.0e-03 480774 1 4.8e-01 4.5e-03 1.8e-02 1 4.7e-01 3.9e-03 1.7e-02 1 4.8e-01 3.9e-03 2.0e-02 1246974 1 1.1e+00 9.4e-03 4.7e-02 1 1.1e+00 9.1e-03 4.1e-02 1 1.2e+00 9.1e-03 5.3e-02 3234251 1 3.2e-01 2.8e-02 9.4e-02 M 1 3.8e-02 3.8e-02 3.8e-02 M 1 1.1e-01 2.8e-02 4.6e-02 M 8388608 1 6.5e-01 6.8e-02 1.7e-01 R 1 1.0e-01 1.0e-01 1.0e-01 R 0 0.0e+00 1.0e+30 0.0e+00 R Explanation: M = only the best method is used R = only best method and only 2 repetitions 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 1.5e+00 1.2e-02 5.5e-02 93 4.6e-01 3.5e-03 1.7e-02 94 4.7e-01 3.6e-03 1.7e-02 2 150 7.6e-01 6.0e-03 2.8e-02 66 3.3e-01 2.6e-03 1.2e-02 66 3.3e-01 2.5e-03 1.2e-02 4 75 3.9e-01 3.0e-03 1.4e-02 64 3.2e-01 2.5e-03 1.2e-02 65 3.3e-01 2.5e-03 1.2e-02 8 62 3.2e-01 2.5e-03 1.1e-02 64 3.2e-01 2.5e-03 1.2e-02 63 3.1e-01 2.5e-03 1.2e-02 16 62 3.3e-01 2.5e-03 1.2e-02 63 3.3e-01 2.5e-03 1.2e-02 63 3.2e-01 2.4e-03 1.2e-02 32 61 3.2e-01 2.5e-03 1.2e-02 63 3.3e-01 2.5e-03 1.2e-02 65 3.4e-01 2.6e-03 1.3e-02 64 62 3.3e-01 2.6e-03 1.2e-02 62 3.3e-01 2.4e-03 1.2e-02 63 3.3e-01 2.5e-03 1.2e-02 128 60 3.3e-01 2.5e-03 1.2e-02 63 3.4e-01 2.6e-03 1.3e-02 63 3.4e-01 2.6e-03 1.3e-02 256 59 3.2e-01 2.4e-03 1.2e-02 61 3.3e-01 2.5e-03 1.2e-02 61 3.4e-01 2.6e-03 1.2e-02 512 60 3.3e-01 2.6e-03 1.2e-02 59 3.2e-01 2.4e-03 1.2e-02 59 3.3e-01 2.5e-03 1.3e-02 1024 58 3.3e-01 2.6e-03 1.2e-02 60 3.4e-01 2.5e-03 1.3e-02 58 3.3e-01 2.5e-03 1.2e-02 2048 55 4.3e-01 2.7e-03 1.5e-02 59 4.6e-01 2.8e-03 1.6e-02 58 4.5e-01 2.8e-03 1.6e-02 4096 51 4.8e-01 3.0e-03 1.7e-02 52 5.2e-01 2.9e-03 1.8e-02 51 4.8e-01 2.9e-03 1.7e-02 10624 32 5.0e-01 2.7e-03 2.3e-02 34 5.2e-01 3.0e-03 1.7e-02 34 5.3e-01 2.9e-03 3.2e-02 27554 23 5.9e-01 3.1e-03 2.1e-02 21 5.3e-01 2.8e-03 1.7e-02 22 5.5e-01 2.8e-03 1.9e-02 71468 14 8.5e-01 3.3e-03 3.0e-02 14 8.4e-01 2.8e-03 2.9e-02 15 8.9e-01 3.5e-03 3.2e-02 185364 8 9.4e-01 3.6e-03 3.7e-02 9 1.0e+00 4.2e-03 3.5e-02 8 9.1e-01 3.0e-03 3.3e-02 480774 4 1.1e+00 3.6e-03 4.8e-02 4 1.1e+00 3.9e-03 4.5e-02 5 1.5e+00 4.7e-03 5.5e-02 1246974 2 1.7e+00 4.7e-03 8.4e-02 1 8.6e-01 2.4e-03 5.1e-02 2 1.7e+00 4.9e-03 8.5e-02 3234251 1 4.5e-01 7.3e-03 5.9e-02 M 1 6.3e-01 7.4e-03 8.0e-02 M 1 3.6e-01 7.2e-03 6.4e-02 M 8388608 1 9.8e-01 1.8e-02 1.4e-01 R 1 1.4e+00 1.8e-02 1.8e-01 R 0 0.0e+00 1.0e+30 0.0e+00 R Explanation: M = only the best method is used R = only best method and only 2 repetitions SECTION-LOOPLNGS-END SECTION-ELAPSED-BEGIN measurment loop: elapsed time = 136.752 sec sum of max elapsed time per entries above = 135.912 sec difference to elapsed time = 0.840 sec = 0.6% sum based on fastest repetition = 123.731 sec difference to elapsed time = 13.021 sec = 9.5% 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-8*2fix 1 16 1.00 1.00 0 ( 0 0 2 ) p01 ring-4*4fix 2 32 2.00 1.00 0 ( 1 0 0 ) p02 ring-2*8fix 2 32 2.00 1.00 0 ( 0 0 0 ) p03 ring-1*16fix 2 32 2.00 1.00 0 ( 0 0 0 ) p04 ring-1*16fix 2 32 2.00 1.00 0 ( 0 2 0 ) p05 ring-1*16fix 2 32 2.00 1.00 0 ( 0 0 0 ) p06 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p07 random-cyc-1dim 2 32 2.00 1.00 0 ( 1 0 0 ) p08 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 0 0 ) p09 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p10 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 1 1 ) p11 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p12 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p13 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 0 ) p14 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 0 0 ) p15 random-cyc-1dim 2 32 2.00 1.00 0 ( 1 0 0 ) p16 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 0 ) p17 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p18 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 0 ) p19 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p20 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p21 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 1 ) p22 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 0 ) p23 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 0 ) p24 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 0 2 ) p25 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 2 ) p26 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 0 ) p27 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 2 ) p28 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p29 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 2 ) p30 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 2 ) p31 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p32 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p33 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 0 ) p34 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p35 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 2 ) p36 worst-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p37 best bi-section 2 16 1.00 0.50 0 ( 2 2 2 ) p38 worst bi-section 2 16 1.00 0.50 0 ( 1 2 1 ) p39 one PingPong Pair 2 2 1.00 0.50 14 ( 0 0 0 ) p40 acyclic-2dim-all 4 48 3.00 0.75 0 ( 2 2 2 ) p41 acyclic-3dim-all 6 56 3.50 0.58 0 ( 2 2 2 ) p42 cyclic-2dim-x 2 32 2.00 1.00 0 ( 2 0 0 ) p43 cyclic-2dim-y 2 32 2.00 1.00 0 ( 1 0 0 ) p44 cyclic-2dim-all 4 64 4.00 1.00 0 ( 2 0 2 ) p45 cyclic-3dim-x 2 32 2.00 1.00 0 ( 2 2 0 ) p46 cyclic-3dim-y 1 16 1.00 1.00 0 ( 2 2 2 ) p47 cyclic-3dim-z 1 16 1.00 1.00 0 ( 0 2 0 ) p48 cyclic-3dim-all 4 64 4.00 1.00 0 ( 2 2 2 ) 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-8*2fix : 37.957 26.250 37.297 -> 37.957 -> 607.305 MByte/s p01 ring-4*4fix : 38.636 29.948 36.327 -> 38.636 -> 618.177 MByte/s p02 ring-2*8fix : 40.574 30.341 36.630 -> 40.574 -> 649.180 MByte/s p03 ring-1*16fix : 40.345 25.110 37.034 -> 40.345 -> 645.526 MByte/s p04 ring-1*16fix : 39.903 25.233 37.136 -> 39.903 -> 638.440 MByte/s p05 ring-1*16fix : 40.404 26.343 37.186 -> 40.404 -> 646.466 MByte/s p06 random-cyc-1dim : 63.496 36.021 60.462 -> 63.496 -> 1015.943 MByte/s p07 random-cyc-1dim : 48.147 28.876 47.155 -> 48.147 -> 770.355 MByte/s p08 random-cyc-1dim : 52.973 30.301 53.125 -> 53.125 -> 850.006 MByte/s p09 random-cyc-1dim : 55.867 31.127 52.607 -> 55.867 -> 893.867 MByte/s p10 random-cyc-1dim : 47.865 27.944 46.187 -> 47.865 -> 765.835 MByte/s p11 random-cyc-1dim : 55.503 35.691 53.579 -> 55.503 -> 888.047 MByte/s p12 random-cyc-1dim : 49.313 30.666 46.458 -> 49.313 -> 789.009 MByte/s p13 random-cyc-1dim : 45.242 27.220 42.442 -> 45.242 -> 723.879 MByte/s p14 random-cyc-1dim : 63.212 39.488 58.087 -> 63.212 -> 1011.397 MByte/s p15 random-cyc-1dim : 49.881 29.370 47.606 -> 49.881 -> 798.098 MByte/s p16 random-cyc-1dim : 42.172 24.988 40.026 -> 42.172 -> 674.751 MByte/s p17 random-cyc-1dim : 67.786 37.392 65.739 -> 67.786 -> 1084.569 MByte/s p18 random-cyc-1dim : 50.451 33.338 48.193 -> 50.451 -> 807.218 MByte/s p19 random-cyc-1dim : 70.092 41.897 62.247 -> 70.092 -> 1121.478 MByte/s p20 random-cyc-1dim : 53.692 30.876 51.525 -> 53.692 -> 859.075 MByte/s p21 random-cyc-1dim : 53.154 31.545 50.001 -> 53.154 -> 850.458 MByte/s p22 random-cyc-1dim : 64.748 40.339 62.353 -> 64.748 -> 1035.963 MByte/s p23 random-cyc-1dim : 55.435 35.210 52.753 -> 55.435 -> 886.961 MByte/s p24 random-cyc-1dim : 54.209 36.473 52.693 -> 54.209 -> 867.350 MByte/s p25 random-cyc-1dim : 56.951 26.529 49.174 -> 56.951 -> 911.215 MByte/s p26 random-cyc-1dim : 80.309 47.629 70.967 -> 80.309 -> 1284.943 MByte/s p27 random-cyc-1dim : 70.385 40.615 65.813 -> 70.385 -> 1126.162 MByte/s p28 random-cyc-1dim : 70.625 41.840 62.192 -> 70.625 -> 1129.992 MByte/s p29 random-cyc-1dim : 49.043 28.006 42.336 -> 49.043 -> 784.695 MByte/s p30 random-cyc-1dim : 65.864 40.214 57.856 -> 65.864 -> 1053.831 MByte/s p31 random-cyc-1dim : 49.278 28.999 47.551 -> 49.278 -> 788.455 MByte/s p32 random-cyc-1dim : 62.490 37.114 59.279 -> 62.490 -> 999.836 MByte/s p33 random-cyc-1dim : 50.914 26.006 49.092 -> 50.914 -> 814.627 MByte/s p34 random-cyc-1dim : 49.662 33.068 46.926 -> 49.662 -> 794.596 MByte/s p35 random-cyc-1dim : 48.557 28.403 45.203 -> 48.557 -> 776.913 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 71.212 40.795 64.332 -> 71.212 -> 1139.399 MByte/s p37 best bi-section : 29.680 26.063 38.573 -> 38.573 -> 617.169 MByte/s p38 worst bi-section : 36.463 26.976 37.401 -> 37.401 -> 598.414 MByte/s p39 one PingPong Pair : 2.331 0.746 0.746 -> 2.331 -> 37.290 MByte/s p40 acyclic-2dim-all : 55.666 51.312 60.021 -> 60.021 -> 960.343 MByte/s p41 acyclic-3dim-all : 58.890 61.159 87.172 -> 87.172 -> 1394.750 MByte/s p42 cyclic-2dim-x : 154.174 70.548 146.568 -> 154.174 -> 2466.787 MByte/s p43 cyclic-2dim-y : 39.136 30.168 35.608 -> 39.136 -> 626.176 MByte/s p44 cyclic-2dim-all : 62.119 54.739 66.519 -> 66.519 -> 1064.307 MByte/s p45 cyclic-3dim-x : 155.854 71.146 146.557 -> 155.854 -> 2493.658 MByte/s p46 cyclic-3dim-y : 156.506 67.423 144.351 -> 156.506 -> 2504.094 MByte/s p47 cyclic-3dim-z : 39.025 26.345 37.852 -> 39.025 -> 624.403 MByte/s p48 cyclic-3dim-all : 92.692 72.322 102.168 -> 102.168 -> 1634.691 MByte/s log_avg of all rings : 39.624 27.123 36.933 || 39.624 -> 633.983 MByte/s log_avg of all random : 55.894 33.109 52.441 || 55.899 -> 894.382 MByte/s log_avg(ring,random) : 47.061 29.967 44.009 ||( 47.063 -> 753.009)MByte/s * size -> accumulated on all pr.: 752.973 479.466 704.150 ||(753.009)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-8*2fix : 36.334 36.953 37.418 -> 37.418 -> 598.695 MByte/s p01 ring-4*4fix : 35.514 38.434 38.521 -> 38.521 -> 616.334 MByte/s p02 ring-2*8fix : 38.771 40.619 39.144 -> 40.619 -> 649.905 MByte/s p03 ring-1*16fix : 38.922 39.986 40.178 -> 40.178 -> 642.853 MByte/s p04 ring-1*16fix : 39.301 36.108 39.977 -> 39.977 -> 639.625 MByte/s p05 ring-1*16fix : 39.754 40.176 40.415 -> 40.415 -> 646.636 MByte/s p06 random-cyc-1dim : 55.532 56.461 59.780 -> 59.780 -> 956.480 MByte/s p07 random-cyc-1dim : 38.310 47.959 44.780 -> 47.959 -> 767.345 MByte/s p08 random-cyc-1dim : 51.469 52.619 52.769 -> 52.769 -> 844.304 MByte/s p09 random-cyc-1dim : 54.038 53.431 56.160 -> 56.160 -> 898.559 MByte/s p10 random-cyc-1dim : 46.085 39.566 45.698 -> 46.085 -> 737.368 MByte/s p11 random-cyc-1dim : 53.417 54.911 55.417 -> 55.417 -> 886.675 MByte/s p12 random-cyc-1dim : 46.033 46.767 48.965 -> 48.965 -> 783.438 MByte/s p13 random-cyc-1dim : 43.194 40.032 45.030 -> 45.030 -> 720.475 MByte/s p14 random-cyc-1dim : 55.703 63.155 60.313 -> 63.155 -> 1010.474 MByte/s p15 random-cyc-1dim : 39.612 46.660 45.584 -> 46.660 -> 746.563 MByte/s p16 random-cyc-1dim : 39.522 41.836 41.860 -> 41.860 -> 669.758 MByte/s p17 random-cyc-1dim : 64.107 65.710 62.629 -> 65.710 -> 1051.362 MByte/s p18 random-cyc-1dim : 48.795 46.855 48.546 -> 48.795 -> 780.726 MByte/s p19 random-cyc-1dim : 61.439 66.891 64.977 -> 66.891 -> 1070.250 MByte/s p20 random-cyc-1dim : 51.862 53.420 52.532 -> 53.420 -> 854.723 MByte/s p21 random-cyc-1dim : 52.806 52.382 50.278 -> 52.806 -> 844.895 MByte/s p22 random-cyc-1dim : 65.351 60.581 61.975 -> 65.351 -> 1045.617 MByte/s p23 random-cyc-1dim : 54.968 52.554 56.231 -> 56.231 -> 899.697 MByte/s p24 random-cyc-1dim : 46.189 53.223 49.875 -> 53.223 -> 851.575 MByte/s p25 random-cyc-1dim : 50.115 50.630 52.850 -> 52.850 -> 845.598 MByte/s p26 random-cyc-1dim : 68.423 77.828 76.904 -> 77.828 -> 1245.245 MByte/s p27 random-cyc-1dim : 70.850 63.605 66.827 -> 70.850 -> 1133.599 MByte/s p28 random-cyc-1dim : 61.899 62.053 66.441 -> 66.441 -> 1063.058 MByte/s p29 random-cyc-1dim : 45.943 44.896 44.896 -> 45.943 -> 735.090 MByte/s p30 random-cyc-1dim : 63.615 64.847 58.460 -> 64.847 -> 1037.552 MByte/s p31 random-cyc-1dim : 48.581 48.403 49.196 -> 49.196 -> 787.143 MByte/s p32 random-cyc-1dim : 57.881 59.953 61.891 -> 61.891 -> 990.258 MByte/s p33 random-cyc-1dim : 49.624 46.881 51.513 -> 51.513 -> 824.215 MByte/s p34 random-cyc-1dim : 48.751 48.453 48.633 -> 48.751 -> 780.009 MByte/s p35 random-cyc-1dim : 45.864 46.877 46.371 -> 46.877 -> 750.036 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 63.291 69.493 71.018 -> 71.018 -> 1136.289 MByte/s p37 best bi-section : 35.412 37.035 35.783 -> 37.035 -> 592.560 MByte/s p38 worst bi-section : 38.860 37.614 37.891 -> 38.860 -> 621.755 MByte/s p39 one PingPong Pair : 2.301 2.224 2.030 -> 2.301 -> 36.822 MByte/s p40 acyclic-2dim-all : 60.072 60.145 58.747 -> 60.145 -> 962.324 MByte/s p41 acyclic-3dim-all : 84.114 84.357 82.572 -> 84.357 -> 1349.707 MByte/s p42 cyclic-2dim-x : 154.887 153.796 154.614 -> 154.887 -> 2478.186 MByte/s p43 cyclic-2dim-y : 37.738 39.217 38.644 -> 39.217 -> 627.465 MByte/s p44 cyclic-2dim-all : 64.691 66.278 65.766 -> 66.278 -> 1060.444 MByte/s p45 cyclic-3dim-x : 153.939 144.055 148.476 -> 153.939 -> 2463.024 MByte/s p46 cyclic-3dim-y : 157.978 153.961 154.245 -> 157.978 -> 2527.649 MByte/s p47 cyclic-3dim-z : 39.356 36.686 39.421 -> 39.421 -> 630.731 MByte/s p48 cyclic-3dim-all : 101.512 102.050 98.202 -> 102.050 -> 1632.792 MByte/s log_avg of all rings : 38.066 38.675 39.261 || 39.504 -> 632.065 MByte/s log_avg of all random : 52.002 52.956 53.658 || 54.784 -> 876.541 MByte/s log_avg(ring,random) : 44.492 45.255 45.899 ||( 46.521 -> 744.332)MByte/s * size -> accumulated on all pr.: 711.865 724.086 734.378 ||(744.332)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-8*2fix p00 method 0 =Sndrcv :( 38.500) 0.026 0.403 6.111 35.310 95.319 119.632 -> 37.957 -> 607.305 MByte/s p00 method 1 =Alltoal :(396.481) 0.003 0.040 0.648 7.656 62.014 119.632 -> 26.250 -> 420.004 MByte/s p00 method 2 =non-blk :( 69.330) 0.014 0.226 3.688 29.463 87.653 119.632 -> 37.297 -> 596.751 MByte/s p01 ring-4*4fix p01 method 0 =Sndrcv :( 35.373) 0.028 0.443 6.661 42.104 91.591 112.289 -> 38.636 -> 618.177 MByte/s p01 method 1 =Alltoal :(195.112) 0.005 0.081 1.278 13.166 72.057 112.289 -> 29.948 -> 479.163 MByte/s p01 method 2 =non-blk :( 56.957) 0.018 0.271 4.130 35.914 96.664 112.289 -> 36.327 -> 581.231 MByte/s p02 ring-2*8fix p02 method 0 =Sndrcv :( 34.857) 0.029 0.445 6.665 42.220 100.305 117.004 -> 40.574 -> 649.180 MByte/s p02 method 1 =Alltoal :(194.947) 0.005 0.081 1.277 12.311 79.899 117.004 -> 30.341 -> 485.452 MByte/s p02 method 2 =non-blk :( 56.091) 0.018 0.275 4.168 36.100 97.649 117.004 -> 36.630 -> 586.085 MByte/s p03 ring-1*16fix p03 method 0 =Sndrcv :( 34.842) 0.029 0.444 6.657 42.551 93.890 119.484 -> 40.345 -> 645.526 MByte/s p03 method 1 =Alltoal :(195.338) 0.005 0.082 1.277 11.486 46.863 119.484 -> 25.110 -> 401.754 MByte/s p03 method 2 =non-blk :( 56.218) 0.018 0.272 4.202 36.431 96.951 119.484 -> 37.034 -> 592.537 MByte/s p04 ring-1*16fix p04 method 0 =Sndrcv :( 34.825) 0.029 0.447 6.684 42.396 94.858 107.571 -> 39.903 -> 638.440 MByte/s p04 method 1 =Alltoal :(197.539) 0.005 0.082 1.280 11.753 53.543 107.571 -> 25.233 -> 403.728 MByte/s p04 method 2 =non-blk :( 56.502) 0.018 0.270 4.152 36.330 98.559 107.571 -> 37.136 -> 594.184 MByte/s p05 ring-1*16fix p05 method 0 =Sndrcv :( 34.805) 0.029 0.448 6.679 42.810 98.255 116.218 -> 40.404 -> 646.466 MByte/s p05 method 1 =Alltoal :(197.958) 0.005 0.081 1.279 11.831 60.038 116.218 -> 26.343 -> 421.492 MByte/s p05 method 2 =non-blk :( 56.280) 0.018 0.271 4.119 36.207 94.856 116.218 -> 37.186 -> 594.980 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 34.102) 0.029 0.455 6.790 44.869 152.563 215.665 -> 63.496 -> 1015.943 MByte/s p06 method 1 =Alltoal :(217.279) 0.005 0.074 1.161 13.662 62.203 215.665 -> 36.021 -> 576.331 MByte/s p06 method 2 =non-blk :( 54.494) 0.018 0.278 4.194 38.005 144.828 215.665 -> 60.462 -> 967.398 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 34.337) 0.029 0.452 6.725 44.297 120.649 144.933 -> 48.147 -> 770.355 MByte/s p07 method 1 =Alltoal :(199.704) 0.005 0.081 1.264 12.375 56.904 144.933 -> 28.876 -> 462.012 MByte/s p07 method 2 =non-blk :( 56.048) 0.018 0.277 4.200 37.074 126.723 144.933 -> 47.155 -> 754.481 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 34.215) 0.029 0.455 6.733 45.213 130.518 180.426 -> 52.973 -> 847.569 MByte/s p08 method 1 =Alltoal :(205.990) 0.005 0.078 1.221 13.179 57.193 180.426 -> 30.301 -> 484.822 MByte/s p08 method 2 =non-blk :( 55.129) 0.018 0.277 4.204 37.875 143.305 180.426 -> 53.125 -> 850.006 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 32.190) 0.031 0.467 7.078 49.885 140.682 175.991 -> 55.867 -> 893.867 MByte/s p09 method 1 =Alltoal :(205.743) 0.005 0.078 1.229 12.360 51.065 175.991 -> 31.127 -> 498.038 MByte/s p09 method 2 =non-blk :( 50.739) 0.020 0.304 4.620 41.826 143.557 175.991 -> 52.607 -> 841.707 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 34.463) 0.029 0.450 6.708 44.922 116.711 147.731 -> 47.865 -> 765.835 MByte/s p10 method 1 =Alltoal :(199.667) 0.005 0.080 1.260 12.480 52.047 147.731 -> 27.944 -> 447.109 MByte/s p10 method 2 =non-blk :( 55.766) 0.018 0.275 4.189 37.534 124.484 147.731 -> 46.187 -> 738.989 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 29.531) 0.034 0.512 7.553 51.945 134.919 173.791 -> 55.503 -> 888.047 MByte/s p11 method 1 =Alltoal :(205.510) 0.005 0.078 1.178 13.395 84.717 173.791 -> 35.691 -> 571.057 MByte/s p11 method 2 =non-blk :( 49.963) 0.020 0.304 4.606 41.638 138.042 173.791 -> 53.579 -> 857.266 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 34.305) 0.029 0.449 6.698 44.496 124.206 149.045 -> 49.313 -> 789.009 MByte/s p12 method 1 =Alltoal :(205.168) 0.005 0.078 1.229 12.716 74.444 149.045 -> 30.666 -> 490.652 MByte/s p12 method 2 =non-blk :( 56.218) 0.018 0.270 4.173 36.771 122.794 149.045 -> 46.458 -> 743.322 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 34.295) 0.029 0.458 6.809 44.832 119.414 129.731 -> 45.242 -> 723.879 MByte/s p13 method 1 =Alltoal :(203.742) 0.005 0.079 1.226 12.808 60.067 129.731 -> 27.220 -> 435.526 MByte/s p13 method 2 =non-blk :( 55.403) 0.018 0.275 4.263 36.918 115.699 129.731 -> 42.442 -> 679.075 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 29.449) 0.034 0.506 7.536 52.261 154.306 169.124 -> 63.212 -> 1011.397 MByte/s p14 method 1 =Alltoal :(211.330) 0.005 0.076 1.193 13.713 88.649 169.124 -> 39.488 -> 631.802 MByte/s p14 method 2 =non-blk :( 50.926) 0.020 0.298 4.623 42.270 153.153 169.124 -> 58.087 -> 929.388 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 34.300) 0.029 0.452 6.739 44.061 123.156 157.205 -> 49.881 -> 798.098 MByte/s p15 method 1 =Alltoal :(198.593) 0.005 0.081 1.258 12.367 50.064 157.205 -> 29.370 -> 469.919 MByte/s p15 method 2 =non-blk :( 56.822) 0.018 0.272 4.231 36.543 124.846 157.205 -> 47.606 -> 761.688 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 34.560) 0.029 0.451 6.704 44.133 100.596 128.865 -> 42.172 -> 674.751 MByte/s p16 method 1 =Alltoal :(199.371) 0.005 0.081 1.264 12.129 48.773 128.865 -> 24.988 -> 399.802 MByte/s p16 method 2 =non-blk :( 56.207) 0.018 0.272 4.216 36.591 109.686 128.865 -> 40.026 -> 640.411 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 28.177) 0.035 0.544 8.110 54.788 182.865 227.321 -> 67.786 -> 1084.569 MByte/s p17 method 1 =Alltoal :(218.385) 0.005 0.073 1.155 14.675 69.113 227.321 -> 37.392 -> 598.277 MByte/s p17 method 2 =non-blk :( 49.207) 0.020 0.308 4.735 43.143 166.153 227.321 -> 65.739 -> 1051.820 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 34.276) 0.029 0.454 6.756 45.233 127.306 138.877 -> 50.451 -> 807.218 MByte/s p18 method 1 =Alltoal :(204.166) 0.005 0.078 1.217 12.958 60.321 138.877 -> 33.338 -> 533.406 MByte/s p18 method 2 =non-blk :( 55.765) 0.018 0.275 4.198 37.612 115.433 138.877 -> 48.193 -> 771.095 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 32.097) 0.031 0.464 7.051 50.752 161.882 224.421 -> 70.092 -> 1121.478 MByte/s p19 method 1 =Alltoal :(217.948) 0.005 0.073 1.118 13.369 85.975 224.421 -> 41.897 -> 670.348 MByte/s p19 method 2 =non-blk :( 50.876) 0.020 0.307 4.698 42.982 153.829 224.421 -> 62.247 -> 995.944 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 29.715) 0.034 0.498 7.444 51.120 139.086 161.636 -> 53.692 -> 859.075 MByte/s p20 method 1 =Alltoal :(203.225) 0.005 0.078 1.227 13.150 62.900 161.636 -> 30.876 -> 494.024 MByte/s p20 method 2 =non-blk :( 51.112) 0.020 0.303 4.632 42.065 140.771 161.636 -> 51.525 -> 824.400 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 29.770) 0.034 0.488 7.545 51.766 129.731 169.760 -> 53.154 -> 850.458 MByte/s p21 method 1 =Alltoal :(204.577) 0.005 0.079 1.238 13.011 64.374 169.760 -> 31.545 -> 504.722 MByte/s p21 method 2 =non-blk :( 51.199) 0.020 0.304 4.601 41.600 149.752 169.760 -> 50.001 -> 800.009 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 28.631) 0.035 0.523 7.810 54.346 176.566 174.739 -> 64.748 -> 1035.963 MByte/s p22 method 1 =Alltoal :(219.648) 0.005 0.073 1.148 14.618 96.924 174.739 -> 40.339 -> 645.416 MByte/s p22 method 2 =non-blk :( 48.973) 0.020 0.316 4.689 43.142 152.211 174.739 -> 62.353 -> 997.648 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 29.314) 0.034 0.503 7.582 51.505 141.253 161.329 -> 55.435 -> 886.961 MByte/s p23 method 1 =Alltoal :(203.444) 0.005 0.078 1.232 13.209 87.561 161.329 -> 35.210 -> 563.362 MByte/s p23 method 2 =non-blk :( 50.367) 0.020 0.304 4.529 38.927 144.484 161.329 -> 52.753 -> 844.042 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 29.438) 0.034 0.504 7.496 51.973 138.607 171.578 -> 54.209 -> 867.350 MByte/s p24 method 1 =Alltoal :(207.331) 0.005 0.077 1.219 13.509 79.231 171.578 -> 36.473 -> 583.567 MByte/s p24 method 2 =non-blk :( 51.893) 0.019 0.298 4.446 41.333 138.558 171.578 -> 52.693 -> 843.080 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 28.292) 0.035 0.545 8.024 53.383 164.080 127.376 -> 56.951 -> 911.215 MByte/s p25 method 1 =Alltoal :(225.070) 0.004 0.072 1.134 14.629 45.578 127.376 -> 26.529 -> 424.467 MByte/s p25 method 2 =non-blk :( 50.000) 0.020 0.313 4.668 41.407 149.661 127.376 -> 49.174 -> 786.782 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 33.893) 0.030 0.459 6.825 45.456 216.899 235.490 -> 80.309 -> 1284.943 MByte/s p26 method 1 =Alltoal :(222.111) 0.005 0.072 1.135 13.703 97.817 235.490 -> 47.629 -> 762.070 MByte/s p26 method 2 =non-blk :( 55.420) 0.018 0.275 4.201 38.496 200.515 235.490 -> 70.967 -> 1135.466 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 29.543) 0.034 0.513 7.584 54.209 183.519 234.168 -> 70.385 -> 1126.162 MByte/s p27 method 1 =Alltoal :(221.537) 0.005 0.073 1.139 13.847 74.414 234.168 -> 40.615 -> 649.843 MByte/s p27 method 2 =non-blk :( 49.118) 0.020 0.318 4.648 42.996 164.458 234.168 -> 65.813 -> 1053.006 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 29.134) 0.034 0.515 7.686 52.667 140.652 240.668 -> 70.625 -> 1129.992 MByte/s p28 method 1 =Alltoal :(226.112) 0.004 0.071 1.122 13.824 80.419 240.668 -> 41.840 -> 669.444 MByte/s p28 method 2 =non-blk :( 49.587) 0.020 0.316 4.653 40.531 127.077 240.668 -> 62.192 -> 995.066 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 34.089) 0.029 0.457 6.767 44.372 114.733 141.297 -> 49.043 -> 784.695 MByte/s p29 method 1 =Alltoal :(206.963) 0.005 0.078 1.219 12.942 58.725 141.297 -> 28.006 -> 448.100 MByte/s p29 method 2 =non-blk :( 56.319) 0.018 0.274 4.179 37.253 116.088 141.297 -> 42.336 -> 677.370 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 28.905) 0.035 0.520 7.785 53.387 156.022 186.248 -> 65.864 -> 1053.831 MByte/s p30 method 1 =Alltoal :(232.129) 0.004 0.069 1.097 14.587 99.657 186.248 -> 40.214 -> 643.430 MByte/s p30 method 2 =non-blk :( 49.532) 0.020 0.308 4.600 42.571 137.937 186.248 -> 57.856 -> 925.701 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 34.348) 0.029 0.453 6.735 45.186 118.154 145.142 -> 49.278 -> 788.455 MByte/s p31 method 1 =Alltoal :(206.485) 0.005 0.078 1.233 12.607 54.979 145.142 -> 28.999 -> 463.977 MByte/s p31 method 2 =non-blk :( 56.075) 0.018 0.276 4.178 37.487 125.354 145.142 -> 47.551 -> 760.811 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 28.939) 0.035 0.521 7.753 52.338 147.082 190.425 -> 62.490 -> 999.836 MByte/s p32 method 1 =Alltoal :(218.643) 0.005 0.074 1.167 13.889 85.245 190.425 -> 37.114 -> 593.817 MByte/s p32 method 2 =non-blk :( 50.324) 0.020 0.307 4.613 42.381 151.868 190.425 -> 59.279 -> 948.468 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 29.742) 0.034 0.506 7.545 50.827 142.017 155.408 -> 50.914 -> 814.627 MByte/s p33 method 1 =Alltoal :(205.888) 0.005 0.078 1.226 13.115 43.421 155.408 -> 26.006 -> 416.092 MByte/s p33 method 2 =non-blk :( 50.919) 0.020 0.298 4.536 41.321 141.260 155.408 -> 49.092 -> 785.468 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 29.844) 0.034 0.497 7.376 50.167 129.494 150.892 -> 49.662 -> 794.596 MByte/s p34 method 1 =Alltoal :(201.835) 0.005 0.079 1.254 12.812 81.586 150.892 -> 33.068 -> 529.091 MByte/s p34 method 2 =non-blk :( 51.339) 0.019 0.297 4.506 38.291 120.582 150.892 -> 46.926 -> 750.810 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 29.312) 0.034 0.511 7.580 52.564 113.907 124.687 -> 48.557 -> 776.913 MByte/s p35 method 1 =Alltoal :(217.778) 0.005 0.073 1.152 14.027 69.725 124.687 -> 28.403 -> 454.456 MByte/s p35 method 2 =non-blk :( 51.382) 0.019 0.297 4.534 41.922 118.823 124.687 -> 45.203 -> 723.245 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 29.423) 0.034 0.501 7.482 54.872 190.171 219.755 -> 71.212 -> 1139.399 MByte/s p36 method 1 =Alltoal :(202.574) 0.005 0.079 1.221 13.645 85.324 219.755 -> 40.795 -> 652.725 MByte/s p36 method 2 =non-blk :( 49.430) 0.020 0.309 4.751 45.691 172.655 219.755 -> 64.332 -> 1029.314 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 29.718) 0.017 0.250 3.612 28.267 78.897 119.326 -> 29.680 -> 474.888 MByte/s p37 method 1 =Alltoal :(196.166) 0.003 0.041 0.646 7.633 63.007 119.326 -> 26.063 -> 417.015 MByte/s p37 method 2 =non-blk :( 34.724) 0.014 0.214 3.539 29.572 103.145 119.326 -> 38.573 -> 617.169 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 28.719) 0.017 0.256 3.817 30.096 95.861 116.859 -> 36.463 -> 583.416 MByte/s p38 method 1 =Alltoal :(197.574) 0.003 0.041 0.624 8.771 48.999 116.859 -> 26.976 -> 431.613 MByte/s p38 method 2 =non-blk :( 33.672) 0.015 0.238 3.652 30.620 103.246 116.859 -> 37.401 -> 598.414 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 29.816) 0.002 0.028 0.502 2.421 4.450 8.027 -> 2.331 -> 37.290 MByte/s p39 method 1 =Alltoal :(-------) 0.000 0.000 0.000 0.000 0.000 8.027 -> 0.746 -> 11.942 MByte/s p39 method 2 =non-blk :(-------) 0.000 0.000 0.000 0.000 0.000 8.027 -> 0.746 -> 11.942 MByte/s p40 acyclic-2dim-all p40 method 0 =Sndrcv :( 22.583) 0.033 0.503 7.413 52.005 133.459 164.034 -> 55.666 -> 890.657 MByte/s p40 method 1 =Alltoal :(100.084) 0.007 0.121 1.826 19.122 143.397 164.034 -> 51.312 -> 820.985 MByte/s p40 method 2 =non-blk :( 41.761) 0.018 0.279 4.200 40.843 184.304 164.034 -> 60.021 -> 960.343 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 18.884) 0.031 0.466 6.873 51.724 132.250 279.208 -> 58.890 -> 942.243 MByte/s p41 method 1 =Alltoal :( 67.777) 0.009 0.139 2.156 23.152 146.650 279.208 -> 61.159 -> 978.537 MByte/s p41 method 2 =non-blk :( 25.959) 0.022 0.349 5.281 54.030 257.821 279.208 -> 87.172 -> 1394.750 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 16.422) 0.061 0.982 13.313 113.623 367.503 462.538 -> 154.174 -> 2466.787 MByte/s p42 method 1 =Alltoal :(196.172) 0.005 0.081 1.279 16.657 106.715 462.538 -> 70.548 -> 1128.763 MByte/s p42 method 2 =non-blk :( 36.973) 0.027 0.433 6.578 75.825 399.819 462.538 -> 146.568 -> 2345.089 MByte/s p43 cyclic-2dim-y p43 method 0 =Sndrcv :( 34.853) 0.029 0.445 6.652 41.662 94.014 114.045 -> 39.136 -> 626.176 MByte/s p43 method 1 =Alltoal :(197.481) 0.005 0.081 1.278 13.145 83.255 114.045 -> 30.168 -> 482.689 MByte/s p43 method 2 =non-blk :( 57.220) 0.017 0.274 4.183 36.041 95.409 114.045 -> 35.608 -> 569.720 MByte/s p44 cyclic-2dim-all p44 method 0 =Sndrcv :( 25.281) 0.040 0.611 8.990 59.687 154.177 205.734 -> 62.119 -> 993.901 MByte/s p44 method 1 =Alltoal :(100.389) 0.010 0.160 2.495 25.137 144.421 205.734 -> 54.739 -> 875.819 MByte/s p44 method 2 =non-blk :( 45.617) 0.022 0.336 5.080 49.338 182.191 205.734 -> 66.519 -> 1064.307 MByte/s p45 cyclic-3dim-x p45 method 0 =Sndrcv :( 16.498) 0.061 0.986 13.556 113.435 378.297 462.271 -> 155.854 -> 2493.658 MByte/s p45 method 1 =Alltoal :(197.753) 0.005 0.081 1.280 16.657 110.598 462.271 -> 71.146 -> 1138.332 MByte/s p45 method 2 =non-blk :( 36.613) 0.027 0.433 6.436 77.073 391.015 462.271 -> 146.557 -> 2344.905 MByte/s p46 cyclic-3dim-y p46 method 0 =Sndrcv :( 16.510) 0.061 0.949 13.403 113.435 379.935 455.134 -> 156.506 -> 2504.094 MByte/s p46 method 1 =Alltoal :(390.119) 0.003 0.041 0.651 9.305 100.957 455.134 -> 67.423 -> 1078.775 MByte/s p46 method 2 =non-blk :( 37.732) 0.027 0.420 6.071 72.582 396.923 455.134 -> 144.351 -> 2309.614 MByte/s p47 cyclic-3dim-z p47 method 0 =Sndrcv :( 38.310) 0.026 0.392 6.070 36.991 91.729 118.523 -> 39.025 -> 624.403 MByte/s p47 method 1 =Alltoal :(393.864) 0.003 0.041 0.646 7.696 63.050 118.523 -> 26.345 -> 421.525 MByte/s p47 method 2 =non-blk :( 69.447) 0.014 0.229 3.690 30.851 102.545 118.523 -> 37.852 -> 605.636 MByte/s p48 cyclic-3dim-all p48 method 0 =Sndrcv :( 20.636) 0.048 0.733 10.799 75.791 242.007 307.734 -> 92.692 -> 1483.079 MByte/s p48 method 1 =Alltoal :(100.326) 0.010 0.160 2.503 26.173 173.844 307.734 -> 72.322 -> 1157.148 MByte/s p48 method 2 =non-blk :( 42.386) 0.024 0.366 5.565 58.894 283.066 307.734 -> 102.168 -> 1634.691 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.028 0.438 6.573 41.139 95.661 115.286 || 39.624 -> 633.983 MByte/s - ring, method 1 = Alltoal: 0.005 0.072 1.141 11.208 61.439 115.286 || 27.123 -> 433.964 MByte/s - ring, method 2 = non-blk: 0.017 0.264 4.072 34.976 95.316 115.286 || 36.933 -> 590.936 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.032 0.485 7.237 49.082 139.391 167.963 || 55.894 -> 894.297 MByte/s - random, method 1 = Alltoal: 0.005 0.076 1.196 13.303 67.562 167.963 || 33.109 -> 529.740 MByte/s - random, method 2 = non-blk: 0.019 0.293 4.441 39.879 137.526 167.963 || 52.441 -> 839.054 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.030 0.461 6.897 44.935 115.474 139.154 || 47.061 -> 752.973 MByte/s - average, method 1 = Alltoal: 0.005 0.074 1.168 12.211 64.428 139.154 || 29.967 -> 479.466 MByte/s - average, method 2 = non-blk: 0.018 0.278 4.253 37.347 114.492 139.154 || 44.009 -> 704.150 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 0.480 7.376 110.347 718.964 1847.585 2226.460 || 752.973 MByte/s - accumulated, mthd 1 = Alltoal: 0.074 1.189 18.694 195.371 1030.844 2226.460 || 479.466 MByte/s - accumulated, mthd 2 = non-blk: 0.289 4.445 68.041 597.549 1831.880 2226.460 || 704.150 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 0.480 0.030 0.028 0.032 0.030 0.005 0.018 2 0.955 0.060 0.056 0.063 0.060 0.009 0.036 4 1.888 0.118 0.111 0.125 0.118 0.019 0.071 8 3.816 0.238 0.224 0.254 0.238 0.037 0.145 16 7.376 0.461 0.438 0.485 0.461 0.074 0.278 32 14.574 0.911 0.865 0.959 0.911 0.148 0.549 64 28.758 1.797 1.713 1.886 1.797 0.296 1.094 128 55.706 3.482 3.326 3.644 3.482 0.587 2.134 256 110.347 6.897 6.573 7.237 6.897 1.168 4.253 512 216.517 13.532 12.884 14.213 13.532 2.328 8.401 1024 420.175 26.261 25.115 27.459 26.261 4.594 16.407 2048 467.191 29.199 27.186 31.362 29.199 7.158 23.202 4096 718.964 44.935 41.139 49.082 44.935 12.211 37.347 10624 934.288 58.393 50.502 67.517 53.726 23.062 58.216 27554 1379.882 86.243 71.774 103.627 77.356 39.033 86.243 71468 1455.944 90.997 73.674 112.393 88.896 57.724 89.027 185364 1887.895 117.993 97.663 142.556 115.474 64.428 114.492 480774 1936.388 121.024 99.850 146.689 119.653 69.235 111.169 1246974 2068.118 129.257 104.250 160.264 128.467 69.127 94.061 3234251 2133.678 133.355 114.919 154.748 133.355 133.355 133.355 8388608 2226.460 139.154 115.286 167.963 139.154 139.154 139.154 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-8*2fix :( 38.500) 0.026 0.403 6.111 35.310 95.319 119.632 -> 39.552 -> 632.831 MByte/s p01 ring-4*4fix :( 35.373) 0.028 0.443 6.661 42.104 96.664 112.289 -> 39.569 -> 633.111 MByte/s p02 ring-2*8fix :( 34.857) 0.029 0.445 6.665 42.220 100.305 117.004 -> 40.823 -> 653.169 MByte/s p03 ring-1*16fix :( 34.842) 0.029 0.444 6.657 42.551 96.951 119.484 -> 40.951 -> 655.213 MByte/s p04 ring-1*16fix :( 34.825) 0.029 0.447 6.684 42.396 98.559 107.571 -> 40.526 -> 648.412 MByte/s p05 ring-1*16fix :( 34.805) 0.029 0.448 6.679 42.810 98.255 116.218 -> 40.974 -> 655.589 MByte/s p06 random-cyc-1dim :( 34.102) 0.029 0.455 6.790 44.869 152.563 215.665 -> 64.985 -> 1039.762 MByte/s p07 random-cyc-1dim :( 34.337) 0.029 0.452 6.725 44.297 126.723 144.933 -> 49.619 -> 793.901 MByte/s p08 random-cyc-1dim :( 34.215) 0.029 0.455 6.733 45.213 143.305 180.426 -> 55.180 -> 882.884 MByte/s p09 random-cyc-1dim :( 32.190) 0.031 0.467 7.078 49.885 143.557 175.991 -> 56.984 -> 911.742 MByte/s p10 random-cyc-1dim :( 34.463) 0.029 0.450 6.708 44.922 124.484 147.731 -> 49.320 -> 789.115 MByte/s p11 random-cyc-1dim :( 29.531) 0.034 0.512 7.553 51.945 138.042 173.791 -> 57.159 -> 914.537 MByte/s p12 random-cyc-1dim :( 34.305) 0.029 0.449 6.698 44.496 124.206 149.045 -> 49.817 -> 797.071 MByte/s p13 random-cyc-1dim :( 34.295) 0.029 0.458 6.809 44.832 119.414 129.731 -> 45.822 -> 733.148 MByte/s p14 random-cyc-1dim :( 29.449) 0.034 0.506 7.536 52.261 154.306 169.124 -> 64.686 -> 1034.974 MByte/s p15 random-cyc-1dim :( 34.300) 0.029 0.452 6.739 44.061 124.846 157.205 -> 50.909 -> 814.546 MByte/s p16 random-cyc-1dim :( 34.560) 0.029 0.451 6.704 44.133 109.686 128.865 -> 43.238 -> 691.803 MByte/s p17 random-cyc-1dim :( 28.177) 0.035 0.544 8.110 54.788 182.865 227.321 -> 70.204 -> 1123.267 MByte/s p18 random-cyc-1dim :( 34.276) 0.029 0.454 6.756 45.233 127.306 138.877 -> 51.485 -> 823.764 MByte/s p19 random-cyc-1dim :( 32.097) 0.031 0.464 7.051 50.752 161.882 224.421 -> 70.369 -> 1125.897 MByte/s p20 random-cyc-1dim :( 29.715) 0.034 0.498 7.444 51.120 140.771 161.636 -> 54.857 -> 877.708 MByte/s p21 random-cyc-1dim :( 29.770) 0.034 0.488 7.545 51.766 149.752 169.760 -> 55.046 -> 880.730 MByte/s p22 random-cyc-1dim :( 28.631) 0.035 0.523 7.810 54.346 176.566 174.739 -> 65.584 -> 1049.338 MByte/s p23 random-cyc-1dim :( 29.314) 0.034 0.503 7.582 51.505 144.484 161.329 -> 56.909 -> 910.538 MByte/s p24 random-cyc-1dim :( 29.438) 0.034 0.504 7.496 51.973 138.607 171.578 -> 56.610 -> 905.762 MByte/s p25 random-cyc-1dim :( 28.292) 0.035 0.545 8.024 53.383 164.080 127.376 -> 57.366 -> 917.848 MByte/s p26 random-cyc-1dim :( 33.893) 0.030 0.459 6.825 45.456 216.899 235.490 -> 80.565 -> 1289.036 MByte/s p27 random-cyc-1dim :( 29.543) 0.034 0.513 7.584 54.209 183.519 234.168 -> 71.985 -> 1151.765 MByte/s p28 random-cyc-1dim :( 29.134) 0.034 0.515 7.686 52.667 140.652 240.668 -> 70.970 -> 1135.513 MByte/s p29 random-cyc-1dim :( 34.089) 0.029 0.457 6.767 44.372 116.088 141.297 -> 49.254 -> 788.060 MByte/s p30 random-cyc-1dim :( 28.905) 0.035 0.520 7.785 53.387 156.022 186.248 -> 66.612 -> 1065.789 MByte/s p31 random-cyc-1dim :( 34.348) 0.029 0.453 6.735 45.186 125.354 145.142 -> 50.349 -> 805.579 MByte/s p32 random-cyc-1dim :( 28.939) 0.035 0.521 7.753 52.338 151.868 190.425 -> 63.742 -> 1019.873 MByte/s p33 random-cyc-1dim :( 29.742) 0.034 0.506 7.545 50.827 142.017 155.408 -> 52.345 -> 837.514 MByte/s p34 random-cyc-1dim :( 29.844) 0.034 0.497 7.376 50.167 129.494 150.892 -> 50.627 -> 810.031 MByte/s p35 random-cyc-1dim :( 29.312) 0.034 0.511 7.580 52.564 118.823 124.687 -> 49.154 -> 786.458 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 29.423) 0.034 0.501 7.482 54.872 190.171 219.755 -> 71.933 -> 1150.929 MByte/s p37 best bi-section :( 29.718) 0.017 0.250 3.612 29.572 103.145 119.326 -> 38.616 -> 617.851 MByte/s p38 worst bi-section :( 28.719) 0.017 0.256 3.817 30.620 103.246 116.859 -> 39.049 -> 624.778 MByte/s p39 one PingPong Pair :( 29.816) 0.002 0.028 0.502 2.421 4.450 8.027 -> 2.331 -> 37.290 MByte/s p40 acyclic-2dim-all :( 22.583) 0.033 0.503 7.413 52.005 184.304 164.034 -> 63.158 -> 1010.525 MByte/s p41 acyclic-3dim-all :( 18.884) 0.031 0.466 6.873 54.030 257.821 279.208 -> 87.756 -> 1404.102 MByte/s p42 cyclic-2dim-x :( 16.422) 0.061 0.982 13.313 113.623 399.819 462.538 -> 156.399 -> 2502.384 MByte/s p43 cyclic-2dim-y :( 34.853) 0.029 0.445 6.652 41.662 95.409 114.045 -> 40.160 -> 642.555 MByte/s p44 cyclic-2dim-all :( 25.281) 0.040 0.611 8.990 59.687 182.191 205.734 -> 68.890 -> 1102.246 MByte/s p45 cyclic-3dim-x :( 16.498) 0.061 0.986 13.556 113.435 391.015 462.271 -> 156.809 -> 2508.949 MByte/s p46 cyclic-3dim-y :( 16.510) 0.061 0.949 13.403 113.435 396.923 455.134 -> 159.602 -> 2553.631 MByte/s p47 cyclic-3dim-z :( 38.310) 0.026 0.392 6.070 36.991 102.545 118.523 -> 40.601 -> 649.612 MByte/s p48 cyclic-3dim-all :( 20.636) 0.048 0.733 10.799 75.791 283.066 307.734 -> 105.526 -> 1688.415 MByte/s log_avg of all rings : 0.028 0.438 6.573 41.139 97.663 115.286 || 40.395 -> 646.313 MByte/s log_avg of all random : 0.032 0.485 7.237 49.082 142.556 167.963 || 57.054 -> 912.864 MByte/s log_avg(ring,random) : 0.030 0.461 6.897 44.935 117.993 139.154 || 48.007 -> 768.112 MByte/s * size -> accumulated on all pr.: 0.480 7.376 110.347 718.964 1887.895 2226.460 || 768.112 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 768.112 MByte/s on 16 processes ( = 48.007 MByte/s * 16 processes) Ping-pong latency: 29.816 microsec Ping-pong bandwidth: 128.435 MByte/s at Lmax= 8.000 MByte (MByte/s=1e6 Byte/s) (MByte=2**20 Byte) system parameters : 16 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 16:41:12 1999 Total execution wall clock time = 138 seconds SECTION-BEFF-END b_eff = 768.112 MB/s = 48.007 * 16 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000