b_eff = 1225.848 MB/s = 51.077 * 24 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 = 24 2-dim-paterns: size = 6 * 4 3-dim-paterns: size = 4 * 3 * 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-12*2fix 1=ring-6*4fix 2=ring-3*8fix 3=ring-1*24fix 4=ring-1*24fix 5=ring-1*24fix 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: 361.996 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 1.3e+00 4.0e-03 9.4e-02 246 1.0e+00 3.1e-03 6.0e-02 258 1.1e+00 3.3e-03 7.0e-02 2 186 7.9e-01 2.5e-03 5.7e-02 197 8.2e-01 2.5e-03 4.8e-02 194 9.1e-01 2.4e-03 8.0e-02 4 188 8.1e-01 2.5e-03 5.1e-02 198 8.7e-01 2.7e-03 5.8e-02 199 8.3e-01 2.6e-03 5.7e-02 8 189 8.1e-01 2.9e-03 4.8e-02 181 7.0e-01 2.6e-03 4.6e-02 189 7.6e-01 2.7e-03 5.0e-02 16 164 9.7e-01 2.4e-03 5.1e-02 172 8.4e-01 2.4e-03 4.9e-02 172 8.4e-01 2.5e-03 5.4e-02 32 174 8.7e-01 2.6e-03 5.8e-02 177 8.9e-01 2.7e-03 5.8e-02 173 8.7e-01 2.6e-03 5.7e-02 64 170 8.6e-01 2.6e-03 5.3e-02 164 8.2e-01 2.5e-03 5.2e-02 167 8.4e-01 2.5e-03 5.6e-02 128 164 8.6e-01 2.7e-03 5.3e-02 164 8.5e-01 2.8e-03 5.5e-02 164 8.5e-01 2.7e-03 6.3e-02 256 152 8.8e-01 3.3e-03 6.0e-02 145 8.2e-01 3.1e-03 5.8e-02 152 9.0e-01 3.2e-03 5.8e-02 512 115 6.7e-01 2.4e-03 4.4e-02 118 6.9e-01 2.6e-03 4.7e-02 120 6.8e-01 2.4e-03 4.3e-02 1024 119 7.0e-01 2.8e-03 4.5e-02 113 6.8e-01 2.4e-03 7.2e-02 123 6.9e-01 2.4e-03 4.5e-02 2048 105 1.1e+00 2.8e-03 8.1e-02 118 1.0e+00 3.3e-03 7.5e-02 126 1.1e+00 3.6e-03 9.5e-02 4096 94 1.1e+00 3.5e-03 9.1e-02 89 9.4e-01 3.3e-03 8.1e-02 87 1.4e+00 3.2e-03 1.0e-01 10624 51 2.3e+00 2.9e-03 1.1e-01 52 2.3e+00 3.0e-03 1.2e-01 52 2.3e+00 2.9e-03 1.1e-01 27554 33 2.9e+00 2.8e-03 1.5e-01 33 2.8e+00 2.8e-03 1.5e-01 34 2.9e+00 2.9e-03 1.5e-01 71468 22 3.4e+00 4.0e-03 1.7e-01 22 3.3e+00 3.7e-03 1.8e-01 22 3.3e+00 3.7e-03 1.8e-01 185364 10 3.0e+00 4.0e-03 1.2e-01 11 3.1e+00 4.0e-03 1.6e-01 11 3.1e+00 4.3e-03 1.7e-01 480774 4 2.4e+00 4.1e-03 1.2e-01 5 2.8e+00 4.4e-03 1.4e-01 4 2.2e+00 3.8e-03 1.0e-01 1246974 1 1.4e+00 2.7e-03 1.3e-01 2 2.3e+00 4.9e-03 9.6e-02 2 2.3e+00 5.1e-03 9.4e-02 3234251 1 1.5e+00 1.4e-02 1.1e-01 M 1 1.8e+00 8.3e-03 1.1e-01 M 1 1.7e+00 8.1e-03 1.1e-01 M 8388608 1 3.5e+00 3.4e-02 2.3e-01 R 1 4.5e+00 1.6e-02 2.1e-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 1.6e+01 2.4e-01 6.9e-01 27 1.4e+00 2.2e-02 5.0e-02 4 2.0e-01 3.0e-03 6.0e-03 2 150 1.0e+01 1.2e-01 1.2e+00 13 6.8e-01 1.0e-02 2.9e-02 3 1.5e-01 2.3e-03 4.7e-03 4 75 4.1e+00 6.0e-02 2.3e-01 6 3.2e-01 4.7e-03 1.9e-02 3 1.5e-01 2.3e-03 4.7e-03 8 37 2.1e+00 3.0e-02 1.5e-01 3 1.4e-01 2.3e-03 3.9e-03 3 1.5e-01 2.3e-03 4.7e-03 16 18 8.8e-01 1.4e-02 2.4e-02 3 1.5e-01 2.2e-03 5.2e-03 3 1.5e-01 2.4e-03 4.2e-03 32 9 4.8e-01 7.2e-03 3.8e-02 3 1.4e-01 2.2e-03 4.2e-03 3 1.5e-01 2.3e-03 4.2e-03 64 4 2.0e-01 3.2e-03 6.3e-03 3 1.5e-01 2.3e-03 4.1e-03 3 1.5e-01 2.4e-03 3.9e-03 128 3 1.5e-01 2.3e-03 4.5e-03 3 1.5e-01 2.4e-03 1.0e-02 3 1.5e-01 2.3e-03 4.5e-03 256 3 1.5e-01 2.3e-03 1.0e-02 3 1.5e-01 2.2e-03 6.7e-03 3 1.4e-01 2.3e-03 4.2e-03 512 3 1.4e-01 2.3e-03 3.9e-03 3 1.5e-01 2.4e-03 1.1e-02 3 1.4e-01 2.2e-03 4.1e-03 1024 3 1.5e-01 2.3e-03 7.6e-03 3 1.5e-01 2.3e-03 6.1e-03 3 1.6e-01 2.3e-03 1.8e-02 2048 3 2.7e-01 2.8e-03 2.2e-02 3 2.1e-01 2.7e-03 6.4e-03 3 2.1e-01 2.6e-03 6.3e-03 4096 2 1.9e-01 1.9e-03 6.9e-03 2 1.9e-01 1.8e-03 5.9e-03 2 2.5e-01 1.9e-03 7.7e-03 10624 1 2.7e-01 9.5e-04 1.7e-02 2 4.9e-01 2.1e-03 1.9e-02 1 2.4e-01 9.8e-04 1.1e-02 27554 2 8.9e-01 2.6e-03 3.7e-02 1 4.6e-01 1.5e-03 2.6e-02 1 4.4e-01 1.4e-03 2.4e-02 71468 1 1.8e+00 3.2e-03 7.8e-01 1 1.0e+00 2.9e-03 3.9e-02 1 1.1e+00 2.7e-03 4.4e-02 185364 1 2.3e+00 5.0e-03 8.8e-02 1 2.3e+00 4.4e-03 7.4e-02 1 2.3e+00 4.5e-03 6.7e-02 480774 1 5.4e+00 6.7e-03 1.6e-01 1 5.8e+00 6.7e-03 1.8e-01 1 5.8e+00 6.5e-03 1.8e-01 1246974 1 1.2e+01 1.2e-02 3.9e-01 1 1.3e+01 1.2e-02 4.3e-01 1 1.1e+01 1.2e-02 3.9e-01 3234251 1 2.9e-02 2.9e-02 2.9e-02 M 1 2.7e-02 2.7e-02 2.7e-02 M 1 2.8e-02 2.8e-02 2.8e-02 M 8388608 1 6.9e-02 6.9e-02 6.9e-02 R 1 6.7e-02 6.7e-02 6.7e-02 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 8.1e-03 5.7e-02 126 5.6e-01 3.3e-03 2.4e-02 136 6.3e-01 3.7e-03 2.6e-02 2 150 7.1e-01 4.1e-03 2.9e-02 95 4.3e-01 2.5e-03 1.8e-02 92 4.2e-01 2.4e-03 1.7e-02 4 91 4.5e-01 2.6e-03 2.0e-02 94 4.5e-01 2.6e-03 1.8e-02 94 4.5e-01 2.6e-03 2.3e-02 8 87 4.1e-01 2.6e-03 1.6e-02 89 4.0e-01 2.4e-03 1.7e-02 90 4.2e-01 2.5e-03 1.9e-02 16 84 4.3e-01 2.4e-03 2.0e-02 91 4.2e-01 2.5e-03 1.8e-02 91 4.3e-01 2.5e-03 2.0e-02 32 86 4.5e-01 2.5e-03 3.2e-02 92 4.3e-01 2.5e-03 1.8e-02 90 4.4e-01 2.5e-03 2.1e-02 64 85 4.4e-01 2.5e-03 3.2e-02 91 5.3e-01 2.6e-03 5.6e-02 89 4.2e-01 2.6e-03 1.8e-02 128 85 4.4e-01 2.6e-03 1.8e-02 87 4.5e-01 2.5e-03 2.7e-02 87 4.4e-01 2.5e-03 3.5e-02 256 82 4.5e-01 2.6e-03 1.9e-02 85 4.5e-01 2.6e-03 1.9e-02 85 4.5e-01 2.6e-03 1.9e-02 512 78 4.4e-01 2.7e-03 1.8e-02 80 4.5e-01 2.5e-03 2.5e-02 82 4.3e-01 2.5e-03 1.8e-02 1024 72 4.0e-01 2.5e-03 1.8e-02 80 4.3e-01 2.5e-03 2.0e-02 83 4.3e-01 2.6e-03 1.9e-02 2048 73 7.4e-01 2.8e-03 2.6e-02 80 7.7e-01 2.9e-03 2.7e-02 81 7.1e-01 3.0e-03 2.8e-02 4096 66 7.4e-01 3.1e-03 3.1e-02 68 7.6e-01 3.2e-03 2.9e-02 67 9.1e-01 3.1e-03 3.6e-02 10624 40 1.1e+00 2.6e-03 4.9e-02 40 1.1e+00 2.6e-03 3.6e-02 41 9.5e-01 2.7e-03 3.1e-02 27554 29 1.4e+00 2.6e-03 5.4e-02 29 1.3e+00 3.4e-03 4.1e-02 29 1.3e+00 3.0e-03 4.1e-02 71468 21 2.7e+00 4.2e-03 2.1e-01 16 1.8e+00 3.2e-03 6.5e-02 18 2.0e+00 3.3e-03 7.0e-02 185364 9 2.1e+00 3.6e-03 1.1e-01 9 1.9e+00 3.7e-03 7.0e-02 10 2.1e+00 3.8e-03 7.8e-02 480774 4 2.2e+00 3.6e-03 7.7e-02 4 2.0e+00 3.4e-03 7.8e-02 5 2.6e+00 4.3e-03 1.1e-01 1246974 2 2.9e+00 4.9e-03 2.8e-01 2 2.6e+00 4.9e-03 1.2e-01 2 2.6e+00 4.8e-03 1.2e-01 3234251 1 1.8e+00 7.1e-03 3.4e-01 M 1 7.7e-01 7.2e-03 1.1e-01 M 1 9.3e-01 6.9e-03 2.1e-01 M 8388608 1 3.3e+00 1.8e-02 2.7e-01 R 1 1.7e+00 1.8e-02 2.3e-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 = 361.996 sec sum of max elapsed time per entries above = 367.848 sec difference to elapsed time = -5.852 sec = 1.6% sum based on fastest repetition = 312.445 sec difference to elapsed time = 49.552 sec = 13.7% 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-12*2fix 1 24 1.00 1.00 0 ( 2 2 2 ) p01 ring-6*4fix 2 48 2.00 1.00 0 ( 0 0 2 ) p02 ring-3*8fix 2 48 2.00 1.00 0 ( 0 2 0 ) p03 ring-1*24fix 2 48 2.00 1.00 0 ( 2 2 0 ) p04 ring-1*24fix 2 48 2.00 1.00 0 ( 0 0 0 ) p05 ring-1*24fix 2 48 2.00 1.00 0 ( 0 0 0 ) p06 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p07 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 0 ) p08 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 2 0 ) p09 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 2 ) p10 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 2 ) p11 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 2 2 ) p12 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p13 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p14 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p15 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 0 ) p16 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 2 ) p17 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 0 ) p18 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 2 2 ) p19 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 2 0 ) p20 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 2 2 ) p21 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 0 ) p22 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p23 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 0 ) p24 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 2 0 ) p25 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 0 ) p26 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p27 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p28 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 0 ) p29 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 0 ) p30 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p31 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p32 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 2 0 ) p33 random-cyc-1dim 2 48 2.00 1.00 0 ( 2 0 0 ) p34 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 2 ) p35 random-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 2 ) p36 worst-cyc-1dim 2 48 2.00 1.00 0 ( 0 0 0 ) p37 best bi-section 2 24 1.00 0.50 0 ( 2 0 2 ) p38 worst bi-section 2 24 1.00 0.50 0 ( 1 1 1 ) p39 one PingPong Pair 2 2 1.00 0.50 22 ( 0 0 0 ) p40 acyclic-2dim-all 4 76 3.17 0.79 0 ( 2 2 2 ) p41 acyclic-3dim-all 6 92 3.83 0.64 0 ( 0 2 2 ) p42 cyclic-2dim-x 2 48 2.00 1.00 0 ( 0 0 0 ) p43 cyclic-2dim-y 2 48 2.00 1.00 0 ( 0 0 0 ) p44 cyclic-2dim-all 4 96 4.00 1.00 0 ( 2 2 2 ) p45 cyclic-3dim-x 2 48 2.00 1.00 0 ( 0 0 0 ) p46 cyclic-3dim-y 2 48 2.00 1.00 0 ( 0 0 0 ) p47 cyclic-3dim-z 1 24 1.00 1.00 0 ( 2 0 0 ) p48 cyclic-3dim-all 5 120 5.00 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-12*2fix : 126.526 45.559 119.608 -> 126.526 -> 3036.624 MByte/s p01 ring-6*4fix : 129.543 46.801 133.441 -> 133.441 -> 3202.575 MByte/s p02 ring-3*8fix : 139.891 46.274 128.979 -> 139.891 -> 3357.383 MByte/s p03 ring-1*24fix : 63.010 23.266 57.087 -> 63.010 -> 1512.234 MByte/s p04 ring-1*24fix : 66.547 24.770 60.123 -> 66.547 -> 1597.140 MByte/s p05 ring-1*24fix : 65.355 22.778 59.548 -> 65.355 -> 1568.521 MByte/s p06 random-cyc-1dim : 25.101 9.815 24.884 -> 25.101 -> 602.425 MByte/s p07 random-cyc-1dim : 26.633 10.295 26.015 -> 26.633 -> 639.180 MByte/s p08 random-cyc-1dim : 22.960 8.413 21.644 -> 22.960 -> 551.031 MByte/s p09 random-cyc-1dim : 22.809 9.158 23.502 -> 23.502 -> 564.040 MByte/s p10 random-cyc-1dim : 26.102 11.314 27.288 -> 27.288 -> 654.923 MByte/s p11 random-cyc-1dim : 23.836 7.815 23.305 -> 23.836 -> 572.066 MByte/s p12 random-cyc-1dim : 24.975 10.119 25.463 -> 25.463 -> 611.119 MByte/s p13 random-cyc-1dim : 24.223 9.919 25.165 -> 25.165 -> 603.959 MByte/s p14 random-cyc-1dim : 26.069 10.349 26.428 -> 26.428 -> 634.262 MByte/s p15 random-cyc-1dim : 27.724 10.380 29.060 -> 29.060 -> 697.429 MByte/s p16 random-cyc-1dim : 23.670 8.937 23.841 -> 23.841 -> 572.189 MByte/s p17 random-cyc-1dim : 21.645 8.008 22.768 -> 22.768 -> 546.422 MByte/s p18 random-cyc-1dim : 26.012 10.068 25.396 -> 26.012 -> 624.278 MByte/s p19 random-cyc-1dim : 23.686 8.599 24.176 -> 24.176 -> 580.215 MByte/s p20 random-cyc-1dim : 23.471 8.030 23.794 -> 23.794 -> 571.057 MByte/s p21 random-cyc-1dim : 25.346 10.356 26.976 -> 26.976 -> 647.416 MByte/s p22 random-cyc-1dim : 30.443 13.389 31.097 -> 31.097 -> 746.329 MByte/s p23 random-cyc-1dim : 23.906 9.235 24.910 -> 24.910 -> 597.836 MByte/s p24 random-cyc-1dim : 23.749 8.951 24.899 -> 24.899 -> 597.578 MByte/s p25 random-cyc-1dim : 24.886 9.304 25.160 -> 25.160 -> 603.848 MByte/s p26 random-cyc-1dim : 24.574 10.017 26.251 -> 26.251 -> 630.025 MByte/s p27 random-cyc-1dim : 27.542 10.950 28.081 -> 28.081 -> 673.944 MByte/s p28 random-cyc-1dim : 24.806 9.951 25.118 -> 25.118 -> 602.824 MByte/s p29 random-cyc-1dim : 25.029 9.661 26.418 -> 26.418 -> 634.036 MByte/s p30 random-cyc-1dim : 26.932 11.846 27.130 -> 27.130 -> 651.124 MByte/s p31 random-cyc-1dim : 25.346 9.986 26.678 -> 26.678 -> 640.270 MByte/s p32 random-cyc-1dim : 23.450 9.378 24.902 -> 24.902 -> 597.648 MByte/s p33 random-cyc-1dim : 24.815 10.140 24.456 -> 24.815 -> 595.553 MByte/s p34 random-cyc-1dim : 23.233 8.794 23.945 -> 23.945 -> 574.681 MByte/s p35 random-cyc-1dim : 23.000 8.821 25.023 -> 25.023 -> 600.547 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 25.837 9.205 26.516 -> 26.516 -> 636.385 MByte/s p37 best bi-section : 87.246 45.998 126.496 -> 126.496 -> 3035.911 MByte/s p38 worst bi-section : 18.449 20.936 28.795 -> 28.795 -> 691.079 MByte/s p39 one PingPong Pair : 2.112 0.634 0.634 -> 2.112 -> 50.691 MByte/s p40 acyclic-2dim-all : 31.613 13.638 41.909 -> 41.909 -> 1005.820 MByte/s p41 acyclic-3dim-all : 26.219 13.628 41.262 -> 41.262 -> 990.288 MByte/s p42 cyclic-2dim-x : 35.540 14.690 38.992 -> 38.992 -> 935.819 MByte/s p43 cyclic-2dim-y : 137.008 47.412 135.732 -> 137.008 -> 3288.201 MByte/s p44 cyclic-2dim-all : 57.631 25.010 74.200 -> 74.200 -> 1780.793 MByte/s p45 cyclic-3dim-x : 30.381 12.133 28.222 -> 30.381 -> 729.137 MByte/s p46 cyclic-3dim-y : 17.495 8.454 17.202 -> 17.495 -> 419.876 MByte/s p47 cyclic-3dim-z : 127.129 44.771 112.550 -> 127.129 -> 3051.091 MByte/s p48 cyclic-3dim-all : 29.596 15.312 42.659 -> 42.659 -> 1023.814 MByte/s log_avg of all rings : 92.548 33.016 86.564 || 93.006 -> 2232.153 MByte/s log_avg of all random : 24.805 9.667 25.391 || 25.520 -> 612.488 MByte/s log_avg(ring,random) : 47.913 17.865 46.882 ||( 48.719 -> 1169.259)MByte/s * size -> accumulated on all pr.: 1149.920 428.772 1125.176 ||(1169.259)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-12*2fix : 106.394 114.491 110.782 -> 114.491 -> 2747.772 MByte/s p01 ring-6*4fix : 130.604 126.674 123.493 -> 130.604 -> 3134.502 MByte/s p02 ring-3*8fix : 107.466 122.852 117.014 -> 122.852 -> 2948.454 MByte/s p03 ring-1*24fix : 46.399 57.939 59.148 -> 59.148 -> 1419.558 MByte/s p04 ring-1*24fix : 52.779 63.922 60.149 -> 63.922 -> 1534.138 MByte/s p05 ring-1*24fix : 55.410 59.955 59.069 -> 59.955 -> 1438.929 MByte/s p06 random-cyc-1dim : 22.330 25.896 25.981 -> 25.981 -> 623.547 MByte/s p07 random-cyc-1dim : 20.534 26.487 26.914 -> 26.914 -> 645.931 MByte/s p08 random-cyc-1dim : 18.909 19.916 24.057 -> 24.057 -> 577.363 MByte/s p09 random-cyc-1dim : 19.404 24.016 20.623 -> 24.016 -> 576.384 MByte/s p10 random-cyc-1dim : 21.289 27.982 24.598 -> 27.982 -> 671.573 MByte/s p11 random-cyc-1dim : 19.657 23.940 23.605 -> 23.940 -> 574.550 MByte/s p12 random-cyc-1dim : 23.210 25.720 25.215 -> 25.720 -> 617.274 MByte/s p13 random-cyc-1dim : 22.070 24.236 24.340 -> 24.340 -> 584.160 MByte/s p14 random-cyc-1dim : 24.729 27.400 27.785 -> 27.785 -> 666.840 MByte/s p15 random-cyc-1dim : 22.892 28.901 28.751 -> 28.901 -> 693.623 MByte/s p16 random-cyc-1dim : 22.712 24.701 24.607 -> 24.701 -> 592.828 MByte/s p17 random-cyc-1dim : 21.452 22.644 23.193 -> 23.193 -> 556.626 MByte/s p18 random-cyc-1dim : 22.132 22.897 26.538 -> 26.538 -> 636.911 MByte/s p19 random-cyc-1dim : 20.596 23.260 24.685 -> 24.685 -> 592.436 MByte/s p20 random-cyc-1dim : 21.616 24.210 24.553 -> 24.553 -> 589.280 MByte/s p21 random-cyc-1dim : 24.263 26.900 26.587 -> 26.900 -> 645.605 MByte/s p22 random-cyc-1dim : 28.209 32.615 29.486 -> 32.615 -> 782.766 MByte/s p23 random-cyc-1dim : 21.525 24.579 24.579 -> 24.579 -> 589.888 MByte/s p24 random-cyc-1dim : 21.960 25.667 24.364 -> 25.667 -> 616.005 MByte/s p25 random-cyc-1dim : 21.756 26.418 24.968 -> 26.418 -> 634.039 MByte/s p26 random-cyc-1dim : 23.909 26.453 26.285 -> 26.453 -> 634.875 MByte/s p27 random-cyc-1dim : 25.945 27.126 26.975 -> 27.126 -> 651.024 MByte/s p28 random-cyc-1dim : 20.124 25.135 24.351 -> 25.135 -> 603.236 MByte/s p29 random-cyc-1dim : 23.423 25.196 24.618 -> 25.196 -> 604.696 MByte/s p30 random-cyc-1dim : 23.944 24.978 27.840 -> 27.840 -> 668.169 MByte/s p31 random-cyc-1dim : 25.608 26.270 26.959 -> 26.959 -> 647.022 MByte/s p32 random-cyc-1dim : 23.758 22.888 24.639 -> 24.639 -> 591.338 MByte/s p33 random-cyc-1dim : 19.844 25.458 23.511 -> 25.458 -> 610.992 MByte/s p34 random-cyc-1dim : 23.786 24.713 22.375 -> 24.713 -> 593.100 MByte/s p35 random-cyc-1dim : 23.014 23.737 24.428 -> 24.428 -> 586.266 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 25.769 27.901 27.051 -> 27.901 -> 669.624 MByte/s p37 best bi-section : 110.013 102.345 121.107 -> 121.107 -> 2906.571 MByte/s p38 worst bi-section : 28.056 27.380 28.884 -> 28.884 -> 693.210 MByte/s p39 one PingPong Pair : 1.766 1.640 1.704 -> 1.766 -> 42.387 MByte/s p40 acyclic-2dim-all : 33.015 37.770 41.579 -> 41.579 -> 997.887 MByte/s p41 acyclic-3dim-all : 36.423 36.973 38.909 -> 38.909 -> 933.814 MByte/s p42 cyclic-2dim-x : 31.151 37.026 40.347 -> 40.347 -> 968.326 MByte/s p43 cyclic-2dim-y : 122.945 132.027 138.603 -> 138.603 -> 3326.464 MByte/s p44 cyclic-2dim-all : 63.315 66.101 74.678 -> 74.678 -> 1792.282 MByte/s p45 cyclic-3dim-x : 26.716 30.258 30.943 -> 30.943 -> 742.629 MByte/s p46 cyclic-3dim-y : 17.807 14.892 16.732 -> 17.807 -> 427.371 MByte/s p47 cyclic-3dim-z : 122.906 120.092 118.120 -> 122.906 -> 2949.739 MByte/s p48 cyclic-3dim-all : 37.817 40.298 40.558 -> 40.558 -> 973.402 MByte/s log_avg of all rings : 76.639 85.681 83.396 || 86.415 -> 2073.959 MByte/s log_avg of all random : 22.394 25.248 25.178 || 25.851 -> 620.421 MByte/s log_avg(ring,random) : 41.428 46.511 45.823 ||( 47.264 -> 1134.341)MByte/s * size -> accumulated on all pr.: 994.267 1116.256 1099.757 ||(1134.341)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-12*2fix p00 method 0 =Sndrcv :( 12.931) 0.077 1.047 11.404 106.189 145.852 446.939 -> 126.526 -> 3036.624 MByte/s p00 method 1 =Alltoal :(1373.261) 0.001 0.013 0.195 4.030 7.368 446.939 -> 45.559 -> 1093.416 MByte/s p00 method 2 =non-blk :( 27.071) 0.037 0.530 7.901 86.020 133.719 446.939 -> 119.608 -> 2870.586 MByte/s p01 ring-6*4fix p01 method 0 =Sndrcv :( 13.551) 0.074 1.023 10.921 109.494 238.913 430.053 -> 129.543 -> 3109.027 MByte/s p01 method 1 =Alltoal :(676.202) 0.001 0.026 0.459 7.512 10.307 430.053 -> 46.801 -> 1123.229 MByte/s p01 method 2 =non-blk :( 25.338) 0.039 0.593 7.920 90.081 227.175 430.053 -> 133.441 -> 3202.575 MByte/s p02 ring-3*8fix p02 method 0 =Sndrcv :( 13.367) 0.075 1.000 11.152 110.840 248.538 453.843 -> 139.891 -> 3357.383 MByte/s p02 method 1 =Alltoal :(691.094) 0.001 0.027 0.448 7.471 8.962 453.843 -> 46.274 -> 1110.585 MByte/s p02 method 2 =non-blk :( 25.778) 0.039 0.599 8.079 90.037 304.570 453.843 -> 128.979 -> 3095.505 MByte/s p03 ring-1*24fix p03 method 0 =Sndrcv :( 46.632) 0.021 0.346 5.204 37.771 171.409 189.380 -> 63.010 -> 1512.234 MByte/s p03 method 1 =Alltoal :(664.883) 0.002 0.025 0.479 8.664 6.384 189.380 -> 23.266 -> 558.388 MByte/s p03 method 2 =non-blk :( 56.872) 0.018 0.292 3.827 33.215 133.404 189.380 -> 57.087 -> 1370.082 MByte/s p04 ring-1*24fix p04 method 0 =Sndrcv :( 44.816) 0.022 0.366 5.376 38.260 140.946 204.040 -> 66.547 -> 1597.140 MByte/s p04 method 1 =Alltoal :(682.357) 0.001 0.026 0.485 8.295 6.190 204.040 -> 24.770 -> 594.486 MByte/s p04 method 2 =non-blk :( 56.570) 0.018 0.285 3.798 32.447 143.329 204.040 -> 60.123 -> 1442.957 MByte/s p05 ring-1*24fix p05 method 0 =Sndrcv :( 49.585) 0.020 0.378 5.423 37.632 167.269 197.286 -> 65.355 -> 1568.521 MByte/s p05 method 1 =Alltoal :(693.500) 0.001 0.026 0.471 8.047 6.180 197.286 -> 22.778 -> 546.671 MByte/s p05 method 2 =non-blk :( 55.228) 0.018 0.280 3.792 33.213 137.475 197.286 -> 59.548 -> 1429.149 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 45.506) 0.022 0.323 4.348 39.327 51.139 79.723 -> 25.101 -> 602.425 MByte/s p06 method 1 =Alltoal :(398.129) 0.003 0.039 0.656 3.070 6.194 79.723 -> 9.815 -> 235.557 MByte/s p06 method 2 =non-blk :( 49.940) 0.020 0.306 4.433 33.892 59.183 79.723 -> 24.884 -> 597.223 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 43.064) 0.023 0.329 4.827 37.985 54.207 76.195 -> 26.633 -> 639.180 MByte/s p07 method 1 =Alltoal :(504.222) 0.002 0.032 0.509 3.958 6.029 76.195 -> 10.295 -> 247.088 MByte/s p07 method 2 =non-blk :( 50.972) 0.020 0.304 4.500 33.219 59.786 76.195 -> 26.015 -> 624.369 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 51.522) 0.019 0.298 3.893 33.617 45.923 64.188 -> 22.960 -> 551.031 MByte/s p08 method 1 =Alltoal :(459.999) 0.002 0.035 0.567 4.023 6.288 64.188 -> 8.413 -> 201.923 MByte/s p08 method 2 =non-blk :( 52.194) 0.019 0.284 3.971 30.346 46.295 64.188 -> 21.644 -> 519.461 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 47.117) 0.021 0.309 4.363 39.209 46.384 77.369 -> 22.809 -> 547.416 MByte/s p09 method 1 =Alltoal :(424.129) 0.002 0.038 0.621 2.897 6.071 77.369 -> 9.158 -> 219.800 MByte/s p09 method 2 =non-blk :( 48.478) 0.021 0.316 4.533 33.235 50.210 77.369 -> 23.502 -> 564.040 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 38.638) 0.026 0.325 4.778 37.830 49.355 85.863 -> 26.102 -> 626.450 MByte/s p10 method 1 =Alltoal :(445.185) 0.002 0.037 0.618 3.893 7.116 85.863 -> 11.314 -> 271.528 MByte/s p10 method 2 =non-blk :( 50.232) 0.020 0.315 4.439 32.574 55.235 85.863 -> 27.288 -> 654.923 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 37.763) 0.026 0.394 5.523 35.942 54.509 51.542 -> 23.836 -> 572.066 MByte/s p11 method 1 =Alltoal :(464.007) 0.002 0.033 0.539 5.260 6.637 51.542 -> 7.815 -> 187.560 MByte/s p11 method 2 =non-blk :( 43.143) 0.023 0.355 4.890 29.057 53.439 51.542 -> 23.305 -> 559.317 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 44.473) 0.022 0.296 4.857 39.385 50.118 89.747 -> 24.975 -> 599.400 MByte/s p12 method 1 =Alltoal :(428.632) 0.002 0.035 0.565 3.147 6.722 89.747 -> 10.119 -> 242.856 MByte/s p12 method 2 =non-blk :( 48.812) 0.020 0.303 4.467 35.861 50.994 89.747 -> 25.463 -> 611.119 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 51.463) 0.019 0.259 3.819 35.191 48.771 87.323 -> 24.223 -> 581.348 MByte/s p13 method 1 =Alltoal :(420.749) 0.002 0.040 0.675 2.985 6.852 87.323 -> 9.919 -> 238.055 MByte/s p13 method 2 =non-blk :( 53.464) 0.019 0.286 4.300 31.244 53.715 87.323 -> 25.165 -> 603.959 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 42.157) 0.024 0.327 4.218 40.476 50.458 84.421 -> 26.069 -> 625.662 MByte/s p14 method 1 =Alltoal :(465.499) 0.002 0.036 0.547 3.451 6.421 84.421 -> 10.349 -> 248.378 MByte/s p14 method 2 =non-blk :( 48.409) 0.021 0.305 4.358 34.966 60.226 84.421 -> 26.428 -> 634.262 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 27.862) 0.036 0.398 5.312 43.489 53.540 77.813 -> 27.724 -> 665.366 MByte/s p15 method 1 =Alltoal :(445.020) 0.002 0.037 0.622 3.072 7.067 77.813 -> 10.380 -> 249.125 MByte/s p15 method 2 =non-blk :( 40.548) 0.025 0.366 5.306 38.253 63.601 77.813 -> 29.060 -> 697.429 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 41.285) 0.024 0.333 4.668 32.831 48.618 69.745 -> 23.670 -> 568.086 MByte/s p16 method 1 =Alltoal :(479.981) 0.002 0.033 0.530 3.499 6.711 69.745 -> 8.937 -> 214.489 MByte/s p16 method 2 =non-blk :( 53.191) 0.019 0.279 4.057 26.113 60.882 69.745 -> 23.841 -> 572.189 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 52.765) 0.019 0.268 3.995 33.531 47.976 63.390 -> 21.645 -> 519.487 MByte/s p17 method 1 =Alltoal :(456.686) 0.002 0.036 0.593 3.073 5.875 63.390 -> 8.008 -> 192.190 MByte/s p17 method 2 =non-blk :( 49.870) 0.020 0.303 4.386 31.774 54.302 63.390 -> 22.768 -> 546.422 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 44.692) 0.022 0.343 4.578 35.195 51.201 72.098 -> 26.012 -> 624.278 MByte/s p18 method 1 =Alltoal :(530.377) 0.002 0.031 0.553 4.913 6.522 72.098 -> 10.068 -> 241.637 MByte/s p18 method 2 =non-blk :( 57.485) 0.017 0.260 3.798 23.556 54.243 72.098 -> 25.396 -> 609.492 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 39.866) 0.025 0.308 4.359 38.250 49.365 60.196 -> 23.686 -> 568.475 MByte/s p19 method 1 =Alltoal :(422.314) 0.002 0.038 0.626 2.861 6.791 60.196 -> 8.599 -> 206.378 MByte/s p19 method 2 =non-blk :( 50.075) 0.020 0.307 4.491 30.572 60.866 60.196 -> 24.176 -> 580.215 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 39.116) 0.026 0.335 4.943 41.705 52.921 66.358 -> 23.471 -> 563.294 MByte/s p20 method 1 =Alltoal :(470.370) 0.002 0.034 0.545 3.054 5.972 66.358 -> 8.030 -> 192.725 MByte/s p20 method 2 =non-blk :( 47.831) 0.021 0.317 4.664 37.186 53.411 66.358 -> 23.794 -> 571.057 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 49.012) 0.020 0.309 4.081 36.251 46.325 82.465 -> 25.346 -> 608.301 MByte/s p21 method 1 =Alltoal :(489.618) 0.002 0.033 0.579 4.049 6.161 82.465 -> 10.356 -> 248.549 MByte/s p21 method 2 =non-blk :( 51.419) 0.019 0.285 3.947 32.924 66.728 82.465 -> 26.976 -> 647.416 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 35.378) 0.028 0.308 5.278 47.821 51.410 111.580 -> 30.443 -> 730.628 MByte/s p22 method 1 =Alltoal :(470.075) 0.002 0.034 0.568 3.897 6.879 111.580 -> 13.389 -> 321.346 MByte/s p22 method 2 =non-blk :( 41.618) 0.024 0.367 5.166 41.950 65.631 111.580 -> 31.097 -> 746.329 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 47.772) 0.021 0.305 4.630 34.341 48.750 67.959 -> 23.906 -> 573.740 MByte/s p23 method 1 =Alltoal :(431.445) 0.002 0.037 0.631 3.700 5.676 67.959 -> 9.235 -> 221.631 MByte/s p23 method 2 =non-blk :( 60.474) 0.017 0.281 4.171 31.602 58.093 67.959 -> 24.910 -> 597.836 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 45.810) 0.022 0.301 4.306 35.479 48.786 83.041 -> 23.749 -> 569.984 MByte/s p24 method 1 =Alltoal :(550.121) 0.002 0.031 0.497 2.972 5.653 83.041 -> 8.951 -> 214.812 MByte/s p24 method 2 =non-blk :( 50.234) 0.020 0.288 4.409 32.280 58.328 83.041 -> 24.899 -> 597.578 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 34.310) 0.029 0.359 4.932 37.865 52.111 77.856 -> 24.886 -> 597.256 MByte/s p25 method 1 =Alltoal :(443.131) 0.002 0.036 0.575 3.436 6.910 77.856 -> 9.304 -> 223.307 MByte/s p25 method 2 =non-blk :( 49.452) 0.020 0.314 4.551 29.031 60.109 77.856 -> 25.160 -> 603.848 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 48.971) 0.020 0.262 3.966 34.662 50.382 83.859 -> 24.574 -> 589.766 MByte/s p26 method 1 =Alltoal :(502.946) 0.002 0.031 0.502 3.412 5.904 83.859 -> 10.017 -> 240.405 MByte/s p26 method 2 =non-blk :( 51.143) 0.020 0.313 4.388 34.034 55.282 83.859 -> 26.251 -> 630.025 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 32.136) 0.031 0.316 4.941 46.662 55.757 78.879 -> 27.542 -> 661.000 MByte/s p27 method 1 =Alltoal :(430.375) 0.002 0.036 0.568 3.418 6.563 78.879 -> 10.950 -> 262.794 MByte/s p27 method 2 =non-blk :( 41.206) 0.024 0.365 5.209 39.779 58.463 78.879 -> 28.081 -> 673.944 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 44.059) 0.023 0.320 4.555 40.029 47.580 79.634 -> 24.806 -> 595.344 MByte/s p28 method 1 =Alltoal :(385.374) 0.003 0.043 0.646 2.844 6.279 79.634 -> 9.951 -> 238.816 MByte/s p28 method 2 =non-blk :( 49.095) 0.020 0.312 4.454 36.881 56.872 79.634 -> 25.118 -> 602.824 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 30.405) 0.033 0.323 5.007 40.288 50.995 76.599 -> 25.029 -> 600.698 MByte/s p29 method 1 =Alltoal :(456.501) 0.002 0.037 0.587 3.903 5.653 76.599 -> 9.661 -> 231.871 MByte/s p29 method 2 =non-blk :( 41.064) 0.024 0.352 5.068 32.342 58.484 76.599 -> 26.418 -> 634.036 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 51.409) 0.019 0.328 4.058 31.756 57.933 85.236 -> 26.932 -> 646.361 MByte/s p30 method 1 =Alltoal :(490.621) 0.002 0.032 0.563 5.060 6.720 85.236 -> 11.846 -> 284.314 MByte/s p30 method 2 =non-blk :( 57.909) 0.017 0.274 3.829 26.989 56.692 85.236 -> 27.130 -> 651.124 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 43.266) 0.023 0.328 4.352 41.292 49.854 83.273 -> 25.346 -> 608.314 MByte/s p31 method 1 =Alltoal :(569.290) 0.002 0.031 0.480 3.383 6.133 83.273 -> 9.986 -> 239.656 MByte/s p31 method 2 =non-blk :( 48.840) 0.020 0.307 4.512 36.942 58.698 83.273 -> 26.678 -> 640.270 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 43.764) 0.023 0.264 3.664 36.617 48.790 68.103 -> 23.450 -> 562.807 MByte/s p32 method 1 =Alltoal :(373.617) 0.003 0.040 0.648 2.797 5.922 68.103 -> 9.378 -> 225.081 MByte/s p32 method 2 =non-blk :( 51.028) 0.020 0.303 4.311 33.788 56.599 68.103 -> 24.902 -> 597.648 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 53.939) 0.019 0.292 3.852 29.095 57.816 88.699 -> 24.815 -> 595.553 MByte/s p33 method 1 =Alltoal :(471.647) 0.002 0.035 0.602 4.994 6.610 88.699 -> 10.140 -> 243.369 MByte/s p33 method 2 =non-blk :( 68.603) 0.015 0.226 3.520 24.618 59.798 88.699 -> 24.456 -> 586.942 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 42.843) 0.023 0.319 4.599 40.914 45.389 68.263 -> 23.233 -> 557.589 MByte/s p34 method 1 =Alltoal :(438.750) 0.002 0.037 0.584 3.403 5.955 68.263 -> 8.794 -> 211.048 MByte/s p34 method 2 =non-blk :( 48.722) 0.021 0.316 4.623 34.263 48.037 68.263 -> 23.945 -> 574.681 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 39.572) 0.025 0.250 4.452 35.921 47.799 67.935 -> 23.000 -> 552.005 MByte/s p35 method 1 =Alltoal :(407.875) 0.002 0.040 0.628 3.467 6.190 67.935 -> 8.821 -> 211.701 MByte/s p35 method 2 =non-blk :( 41.560) 0.024 0.357 5.112 35.794 51.766 67.935 -> 25.023 -> 600.547 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 57.640) 0.017 0.400 6.068 39.644 55.721 74.379 -> 25.837 -> 620.087 MByte/s p36 method 1 =Alltoal :(616.997) 0.002 0.025 0.426 3.438 5.715 74.379 -> 9.205 -> 220.924 MByte/s p36 method 2 =non-blk :( 50.385) 0.020 0.298 4.237 32.495 62.549 74.379 -> 26.516 -> 636.385 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 14.807) 0.034 0.473 4.184 56.563 76.643 457.793 -> 87.246 -> 2093.912 MByte/s p37 method 1 =Alltoal :(540.455) 0.001 0.015 0.251 4.076 7.358 457.793 -> 45.998 -> 1103.943 MByte/s p37 method 2 =non-blk :( 13.955) 0.036 0.545 7.910 85.733 146.556 457.793 -> 126.496 -> 3035.911 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 47.639) 0.010 0.125 1.656 20.730 26.857 100.409 -> 18.449 -> 442.784 MByte/s p38 method 1 =Alltoal :(553.981) 0.001 0.015 0.229 3.452 33.178 100.409 -> 20.936 -> 502.457 MByte/s p38 method 2 =non-blk :( 32.478) 0.015 0.249 3.387 29.493 65.534 100.409 -> 28.795 -> 691.079 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 27.604) 0.002 0.020 0.189 2.209 5.678 6.001 -> 2.112 -> 50.691 MByte/s p39 method 1 =Alltoal :(-------) 0.000 0.000 0.000 0.000 0.000 6.001 -> 0.634 -> 15.211 MByte/s p39 method 2 =non-blk :(-------) 0.000 0.000 0.000 0.000 0.000 6.001 -> 0.634 -> 15.211 MByte/s p40 acyclic-2dim-all p40 method 0 =Sndrcv :( 23.852) 0.033 0.413 5.494 30.779 74.042 79.866 -> 31.613 -> 758.722 MByte/s p40 method 1 =Alltoal :(278.037) 0.003 0.049 0.826 9.878 13.382 79.866 -> 13.638 -> 327.314 MByte/s p40 method 2 =non-blk :( 30.566) 0.026 0.399 4.689 38.422 121.088 79.866 -> 41.909 -> 1005.820 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 40.949) 0.016 0.214 2.575 17.149 53.907 76.646 -> 26.219 -> 629.245 MByte/s p41 method 1 =Alltoal :(190.755) 0.003 0.057 0.939 6.927 14.411 76.646 -> 13.628 -> 327.068 MByte/s p41 method 2 =non-blk :( 26.376) 0.024 0.374 4.926 41.504 103.374 76.646 -> 41.262 -> 990.288 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 37.305) 0.027 0.300 4.764 44.687 67.548 136.552 -> 35.540 -> 852.961 MByte/s p42 method 1 =Alltoal :(548.563) 0.002 0.030 0.500 4.484 6.714 136.552 -> 14.690 -> 352.555 MByte/s p42 method 2 =non-blk :( 42.369) 0.024 0.372 5.109 42.163 96.601 136.552 -> 38.992 -> 935.819 MByte/s p43 cyclic-2dim-y p43 method 0 =Sndrcv :( 13.577) 0.074 1.037 11.062 109.622 283.528 460.242 -> 137.008 -> 3288.201 MByte/s p43 method 1 =Alltoal :(506.744) 0.002 0.031 0.497 8.043 9.621 460.242 -> 47.412 -> 1137.892 MByte/s p43 method 2 =non-blk :( 26.472) 0.038 0.585 7.962 90.506 261.408 460.242 -> 135.732 -> 3257.557 MByte/s p44 cyclic-2dim-all p44 method 0 =Sndrcv :( 25.220) 0.040 0.491 6.611 62.409 109.463 233.866 -> 57.631 -> 1383.143 MByte/s p44 method 1 =Alltoal :(256.874) 0.004 0.060 0.956 9.432 12.231 233.866 -> 25.010 -> 600.249 MByte/s p44 method 2 =non-blk :( 30.485) 0.033 0.512 6.770 65.676 186.384 233.866 -> 74.200 -> 1780.793 MByte/s p45 cyclic-3dim-x p45 method 0 =Sndrcv :( 41.370) 0.024 0.334 4.319 42.785 58.253 116.076 -> 30.381 -> 729.137 MByte/s p45 method 1 =Alltoal :(542.507) 0.002 0.029 0.468 2.735 6.275 116.076 -> 12.133 -> 291.193 MByte/s p45 method 2 =non-blk :( 59.385) 0.017 0.266 3.868 29.729 59.819 116.076 -> 28.222 -> 677.339 MByte/s p46 cyclic-3dim-y p46 method 0 =Sndrcv :( 54.816) 0.018 0.295 3.565 23.278 38.774 38.808 -> 17.495 -> 419.876 MByte/s p46 method 1 =Alltoal :(611.352) 0.002 0.029 0.478 7.915 11.144 38.808 -> 8.454 -> 202.891 MByte/s p46 method 2 =non-blk :( 81.717) 0.012 0.188 2.840 16.955 35.192 38.808 -> 17.202 -> 412.838 MByte/s p47 cyclic-3dim-z p47 method 0 =Sndrcv :( 13.104) 0.076 1.025 11.198 104.813 169.634 459.749 -> 127.129 -> 3051.091 MByte/s p47 method 1 =Alltoal :(1089.990) 0.001 0.016 0.249 4.454 7.077 459.749 -> 44.771 -> 1074.493 MByte/s p47 method 2 =non-blk :( 27.929) 0.036 0.553 7.847 85.414 211.576 459.749 -> 112.550 -> 2701.199 MByte/s p48 cyclic-3dim-all p48 method 0 =Sndrcv :( 36.734) 0.027 0.294 4.135 35.008 66.338 105.373 -> 29.596 -> 710.295 MByte/s p48 method 1 =Alltoal :(234.607) 0.004 0.075 1.211 8.325 18.714 105.373 -> 15.312 -> 367.482 MByte/s p48 method 2 =non-blk :( 37.840) 0.026 0.405 5.603 47.171 112.844 105.373 -> 42.659 -> 1023.814 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.040 0.609 7.714 64.210 180.860 295.441 || 92.548 -> 2221.151 MByte/s - ring, method 1 = Alltoal: 0.001 0.023 0.404 7.125 7.415 295.441 || 33.016 -> 792.382 MByte/s - ring, method 2 = non-blk: 0.026 0.405 5.506 54.064 170.214 295.441 || 86.564 -> 2077.526 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.024 0.313 4.481 37.548 50.373 76.167 || 24.805 -> 595.329 MByte/s - random, method 1 = Alltoal: 0.002 0.035 0.579 3.538 6.335 76.167 || 9.667 -> 232.016 MByte/s - random, method 2 = non-blk: 0.020 0.307 4.442 32.561 56.671 76.167 || 25.391 -> 609.389 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.031 0.436 5.879 49.102 95.448 150.010 || 47.913 -> 1149.920 MByte/s - average, method 1 = Alltoal: 0.002 0.029 0.484 5.021 6.854 150.010 || 17.865 -> 428.772 MByte/s - average, method 2 = non-blk: 0.023 0.352 4.945 41.957 98.215 150.010 || 46.882 -> 1125.176 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 0.737 10.474 141.103 1178.446 2290.762 3600.235 || 1149.920 MByte/s - accumulated, mthd 1 = Alltoal: 0.041 0.686 11.611 120.495 164.491 3600.235 || 428.772 MByte/s - accumulated, mthd 2 = non-blk: 0.551 8.460 118.691 1006.972 2357.150 3600.235 || 1125.176 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 0.738 0.031 0.040 0.024 0.031 0.002 0.023 2 1.452 0.060 0.077 0.048 0.060 0.003 0.046 4 2.935 0.122 0.158 0.095 0.122 0.007 0.089 8 6.312 0.263 0.335 0.206 0.263 0.015 0.182 16 10.712 0.446 0.609 0.327 0.436 0.029 0.352 32 21.065 0.878 1.182 0.652 0.860 0.058 0.700 64 41.111 1.713 2.280 1.287 1.684 0.115 1.386 128 82.487 3.437 4.611 2.562 3.391 0.230 2.716 256 143.197 5.967 7.714 4.615 5.879 0.484 4.945 512 278.369 11.599 14.438 9.318 11.435 0.987 9.772 1024 585.438 24.393 31.925 18.638 24.079 1.977 19.951 2048 751.211 31.300 40.369 24.269 31.105 3.008 26.283 4096 1178.446 49.102 64.210 37.548 49.102 5.021 41.957 10624 1479.751 61.656 112.660 33.743 44.491 6.008 58.800 27554 2057.496 85.729 164.379 44.710 60.662 7.383 85.339 71468 2245.248 93.552 192.302 45.512 81.758 7.090 90.332 185364 2476.427 103.184 187.616 56.749 95.448 6.854 98.215 480774 2753.901 114.746 223.167 58.999 109.534 7.431 103.017 1246974 3703.667 154.319 339.486 70.149 152.683 9.612 110.009 3234251 3841.740 160.072 321.455 79.710 160.072 160.072 160.072 8388608 3600.235 150.010 295.441 76.167 150.010 150.010 150.010 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-12*2fix :( 12.931) 0.077 1.047 11.404 106.189 145.852 446.939 -> 131.993 -> 3167.839 MByte/s p01 ring-6*4fix :( 13.551) 0.074 1.023 10.921 109.494 238.913 430.053 -> 137.327 -> 3295.856 MByte/s p02 ring-3*8fix :( 13.367) 0.075 1.000 11.152 110.840 304.570 453.843 -> 143.045 -> 3433.090 MByte/s p03 ring-1*24fix :( 46.632) 0.021 0.346 5.204 37.771 171.409 189.380 -> 65.294 -> 1567.051 MByte/s p04 ring-1*24fix :( 44.816) 0.022 0.366 5.376 38.260 143.329 204.040 -> 70.207 -> 1684.965 MByte/s p05 ring-1*24fix :( 49.585) 0.020 0.378 5.423 37.632 167.269 197.286 -> 68.002 -> 1632.047 MByte/s p06 random-cyc-1dim :( 45.506) 0.022 0.323 4.433 39.327 59.183 79.723 -> 27.020 -> 648.471 MByte/s p07 random-cyc-1dim :( 43.064) 0.023 0.329 4.827 37.985 59.786 76.195 -> 28.590 -> 686.157 MByte/s p08 random-cyc-1dim :( 51.522) 0.019 0.298 3.971 33.617 46.295 64.188 -> 24.325 -> 583.793 MByte/s p09 random-cyc-1dim :( 47.117) 0.021 0.316 4.533 39.209 50.210 77.369 -> 24.647 -> 591.520 MByte/s p10 random-cyc-1dim :( 38.638) 0.026 0.325 4.778 37.830 55.235 85.863 -> 28.815 -> 691.559 MByte/s p11 random-cyc-1dim :( 37.763) 0.026 0.394 5.523 35.942 54.509 51.542 -> 25.523 -> 612.556 MByte/s p12 random-cyc-1dim :( 44.473) 0.022 0.303 4.857 39.385 50.994 89.747 -> 26.378 -> 633.077 MByte/s p13 random-cyc-1dim :( 51.463) 0.019 0.286 4.300 35.191 53.715 87.323 -> 26.071 -> 625.716 MByte/s p14 random-cyc-1dim :( 42.157) 0.024 0.327 4.358 40.476 60.226 84.421 -> 28.609 -> 686.621 MByte/s p15 random-cyc-1dim :( 27.862) 0.036 0.398 5.312 43.489 63.601 77.813 -> 30.229 -> 725.487 MByte/s p16 random-cyc-1dim :( 41.285) 0.024 0.333 4.668 32.831 60.882 69.745 -> 26.218 -> 629.235 MByte/s p17 random-cyc-1dim :( 49.870) 0.020 0.303 4.386 33.531 54.302 63.390 -> 23.949 -> 574.772 MByte/s p18 random-cyc-1dim :( 44.692) 0.022 0.343 4.578 35.195 54.243 72.098 -> 27.487 -> 659.683 MByte/s p19 random-cyc-1dim :( 39.866) 0.025 0.308 4.491 38.250 60.866 60.196 -> 25.892 -> 621.408 MByte/s p20 random-cyc-1dim :( 39.116) 0.026 0.335 4.943 41.705 53.411 66.358 -> 25.481 -> 611.534 MByte/s p21 random-cyc-1dim :( 49.012) 0.020 0.309 4.081 36.251 66.728 82.465 -> 28.685 -> 688.428 MByte/s p22 random-cyc-1dim :( 35.378) 0.028 0.367 5.278 47.821 65.631 111.580 -> 32.910 -> 789.846 MByte/s p23 random-cyc-1dim :( 47.772) 0.021 0.305 4.630 34.341 58.093 67.959 -> 26.311 -> 631.457 MByte/s p24 random-cyc-1dim :( 45.810) 0.022 0.301 4.409 35.479 58.328 83.041 -> 26.109 -> 626.615 MByte/s p25 random-cyc-1dim :( 34.310) 0.029 0.359 4.932 37.865 60.109 77.856 -> 27.218 -> 653.222 MByte/s p26 random-cyc-1dim :( 48.971) 0.020 0.313 4.388 34.662 55.282 83.859 -> 27.032 -> 648.778 MByte/s p27 random-cyc-1dim :( 32.136) 0.031 0.365 5.209 46.662 58.463 78.879 -> 29.795 -> 715.090 MByte/s p28 random-cyc-1dim :( 44.059) 0.023 0.320 4.555 40.029 56.872 79.634 -> 27.190 -> 652.550 MByte/s p29 random-cyc-1dim :( 30.405) 0.033 0.352 5.068 40.288 58.484 76.599 -> 27.570 -> 661.680 MByte/s p30 random-cyc-1dim :( 51.409) 0.019 0.328 4.058 31.756 57.933 85.236 -> 28.955 -> 694.915 MByte/s p31 random-cyc-1dim :( 43.266) 0.023 0.328 4.512 41.292 58.698 83.273 -> 28.080 -> 673.921 MByte/s p32 random-cyc-1dim :( 43.764) 0.023 0.303 4.311 36.617 56.599 68.103 -> 25.960 -> 623.035 MByte/s p33 random-cyc-1dim :( 53.939) 0.019 0.292 3.852 29.095 59.798 88.699 -> 26.766 -> 642.383 MByte/s p34 random-cyc-1dim :( 42.843) 0.023 0.319 4.623 40.914 48.037 68.263 -> 25.345 -> 608.291 MByte/s p35 random-cyc-1dim :( 39.572) 0.025 0.357 5.112 35.921 51.766 67.935 -> 25.659 -> 615.822 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 50.385) 0.020 0.400 6.068 39.644 62.549 74.379 -> 28.452 -> 682.839 MByte/s p37 best bi-section :( 13.955) 0.036 0.545 7.910 85.733 146.556 457.793 -> 126.496 -> 3035.911 MByte/s p38 worst bi-section :( 32.478) 0.015 0.249 3.387 29.493 65.534 100.409 -> 29.881 -> 717.134 MByte/s p39 one PingPong Pair :( 27.604) 0.002 0.020 0.189 2.209 5.678 6.001 -> 2.112 -> 50.691 MByte/s p40 acyclic-2dim-all :( 23.852) 0.033 0.413 5.494 38.422 121.088 79.866 -> 42.342 -> 1016.217 MByte/s p41 acyclic-3dim-all :( 26.376) 0.024 0.374 4.926 41.504 103.374 76.646 -> 41.262 -> 990.288 MByte/s p42 cyclic-2dim-x :( 37.305) 0.027 0.372 5.109 44.687 96.601 136.552 -> 41.670 -> 1000.086 MByte/s p43 cyclic-2dim-y :( 13.577) 0.074 1.037 11.062 109.622 283.528 460.242 -> 140.730 -> 3377.518 MByte/s p44 cyclic-2dim-all :( 25.220) 0.040 0.512 6.770 65.676 186.384 233.866 -> 75.213 -> 1805.110 MByte/s p45 cyclic-3dim-x :( 41.370) 0.024 0.334 4.319 42.785 59.819 116.076 -> 32.040 -> 768.968 MByte/s p46 cyclic-3dim-y :( 54.816) 0.018 0.295 3.565 23.278 38.774 38.808 -> 18.313 -> 439.520 MByte/s p47 cyclic-3dim-z :( 13.104) 0.076 1.025 11.198 104.813 211.576 459.749 -> 130.885 -> 3141.241 MByte/s p48 cyclic-3dim-all :( 36.734) 0.027 0.405 5.603 47.171 112.844 105.373 -> 42.659 -> 1023.817 MByte/s log_avg of all rings : 0.040 0.609 7.714 64.210 187.616 295.441 || 96.515 -> 2316.348 MByte/s log_avg of all random : 0.024 0.327 4.615 37.548 56.749 76.167 || 27.031 -> 648.738 MByte/s log_avg(ring,random) : 0.031 0.446 5.967 49.102 103.184 150.010 || 51.077 -> 1225.848 MByte/s * size -> accumulated on all pr.: 0.738 10.712 143.197 1178.446 2476.427 3600.235 || 1225.848 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 1225.848 MByte/s on 24 processes ( = 51.077 MByte/s * 24 processes) Ping-pong latency: 27.604 microsec Ping-pong bandwidth: 144.019 MByte/s at Lmax= 8.000 MByte (MByte/s=1e6 Byte/s) (MByte=2**20 Byte) system parameters : 24 nodes, 1024 MB/node system name : HI-UX/MPP hostname : himiko OS release : 03-00 OS version : ad2b0 machine : SR8000 Date of measurement: Tue Nov 30 16:14:32 1999 Total execution wall clock time = 363 seconds SECTION-BEFF-END b_eff = 1225.848 MB/s = 51.077 * 24 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000