b_eff = 766.605 MB/s = 127.768 * 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: 98.622 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 7.5e-01 4.6e-03 2.9e-02 243 6.0e-01 3.7e-03 2.3e-02 244 6.0e-01 3.7e-03 2.3e-02 2 164 4.1e-01 2.5e-03 1.6e-02 165 4.1e-01 2.5e-03 1.6e-02 165 4.1e-01 2.5e-03 1.6e-02 4 163 4.1e-01 2.5e-03 1.6e-02 163 4.1e-01 2.5e-03 1.6e-02 163 4.1e-01 2.5e-03 1.6e-02 8 161 4.0e-01 2.5e-03 1.6e-02 161 4.0e-01 2.5e-03 1.6e-02 162 4.0e-01 2.5e-03 1.6e-02 16 162 4.3e-01 2.5e-03 1.7e-02 163 4.2e-01 2.5e-03 1.7e-02 164 4.3e-01 2.5e-03 1.7e-02 32 161 4.3e-01 2.6e-03 1.7e-02 162 4.3e-01 2.5e-03 1.7e-02 161 4.2e-01 2.5e-03 1.7e-02 64 156 4.2e-01 2.5e-03 1.7e-02 159 4.3e-01 2.6e-03 1.7e-02 159 4.3e-01 2.6e-03 1.7e-02 128 153 4.3e-01 2.7e-03 1.7e-02 153 4.3e-01 2.7e-03 1.7e-02 154 4.3e-01 2.7e-03 1.7e-02 256 142 4.1e-01 2.4e-03 1.6e-02 144 4.1e-01 2.5e-03 1.6e-02 144 4.2e-01 2.5e-03 1.6e-02 512 146 4.3e-01 2.6e-03 1.7e-02 145 4.2e-01 2.5e-03 1.6e-02 146 4.2e-01 2.6e-03 1.7e-02 1024 140 4.2e-01 2.6e-03 1.7e-02 142 4.2e-01 2.6e-03 1.7e-02 142 4.2e-01 2.6e-03 1.6e-02 2048 136 6.4e-01 3.3e-03 2.4e-02 134 6.3e-01 3.3e-03 2.4e-02 135 6.3e-01 3.3e-03 2.4e-02 4096 103 5.9e-01 3.3e-03 2.2e-02 101 6.0e-01 3.2e-03 2.3e-02 101 6.0e-01 3.2e-03 2.3e-02 10624 60 6.4e-01 3.0e-03 2.1e-02 61 6.5e-01 3.0e-03 2.1e-02 61 6.5e-01 3.1e-03 2.1e-02 27554 38 6.8e-01 3.0e-03 2.1e-02 38 6.8e-01 3.1e-03 2.1e-02 38 6.8e-01 3.1e-03 2.1e-02 71468 24 8.1e-01 3.9e-03 2.5e-02 23 7.7e-01 3.8e-03 2.3e-02 23 7.6e-01 3.6e-03 2.4e-02 185364 11 7.2e-01 4.7e-03 2.2e-02 11 7.1e-01 4.5e-03 2.2e-02 12 7.9e-01 5.1e-03 2.4e-02 480774 4 5.4e-01 3.8e-03 1.8e-02 4 5.3e-01 3.8e-03 1.7e-02 4 5.3e-01 4.0e-03 1.7e-02 1246974 2 6.5e-01 4.4e-03 2.1e-02 2 6.2e-01 4.4e-03 2.2e-02 1 3.0e-01 2.3e-03 1.0e-02 3234251 1 8.1e-01 5.9e-03 2.6e-02 1 7.7e-01 5.9e-03 2.5e-02 1 7.9e-01 5.9e-03 2.6e-02 8388608 1 1.9e+00 1.4e-02 6.2e-02 1 1.9e+00 1.4e-02 6.3e-02 1 1.9e+00 1.4e-02 6.0e-02 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 2.0e+00 3.9e-02 4.6e-02 28 1.9e-01 3.5e-03 4.2e-03 28 1.9e-01 3.5e-03 4.2e-03 2 150 1.0e+00 1.9e-02 2.3e-02 19 1.3e-01 2.5e-03 2.9e-03 19 1.3e-01 2.4e-03 2.8e-03 4 75 5.1e-01 9.8e-03 1.1e-02 19 1.3e-01 2.5e-03 2.9e-03 19 1.3e-01 2.4e-03 2.9e-03 8 37 2.5e-01 4.8e-03 5.6e-03 19 1.3e-01 2.4e-03 2.9e-03 19 1.3e-01 2.4e-03 2.9e-03 16 19 1.3e-01 2.5e-03 3.0e-03 19 1.3e-01 2.5e-03 2.9e-03 19 1.3e-01 2.4e-03 2.9e-03 32 18 1.2e-01 2.4e-03 2.8e-03 19 1.3e-01 2.6e-03 3.0e-03 19 1.3e-01 2.5e-03 3.0e-03 64 19 1.3e-01 2.5e-03 3.0e-03 18 1.2e-01 2.4e-03 2.9e-03 19 1.3e-01 2.5e-03 3.0e-03 128 18 1.3e-01 2.4e-03 3.0e-03 18 1.2e-01 2.4e-03 2.8e-03 19 1.3e-01 2.5e-03 3.0e-03 256 18 1.3e-01 2.5e-03 3.0e-03 18 1.3e-01 2.4e-03 2.9e-03 18 1.3e-01 2.5e-03 3.0e-03 512 18 1.3e-01 2.5e-03 3.0e-03 18 1.3e-01 2.5e-03 3.0e-03 18 1.3e-01 2.5e-03 3.1e-03 1024 17 1.2e-01 2.4e-03 2.9e-03 17 1.2e-01 2.3e-03 2.9e-03 17 1.2e-01 2.4e-03 3.0e-03 2048 17 1.5e-01 2.9e-03 4.0e-03 18 1.6e-01 3.1e-03 4.3e-03 18 1.6e-01 3.0e-03 4.3e-03 4096 14 1.4e-01 2.5e-03 3.8e-03 14 1.4e-01 2.5e-03 3.9e-03 14 1.4e-01 2.5e-03 4.0e-03 10624 10 1.5e-01 2.3e-03 4.2e-03 10 1.4e-01 2.2e-03 4.2e-03 10 1.5e-01 2.2e-03 4.3e-03 27554 8 1.7e-01 2.3e-03 5.1e-03 8 1.7e-01 2.3e-03 5.2e-03 8 1.7e-01 2.3e-03 5.2e-03 71468 6 2.2e-01 2.8e-03 6.7e-03 6 2.2e-01 2.7e-03 6.9e-03 6 2.2e-01 2.7e-03 6.9e-03 185364 4 2.9e-01 2.9e-03 9.6e-03 4 2.8e-01 2.9e-03 9.7e-03 4 2.8e-01 3.0e-03 9.5e-03 480774 2 3.1e-01 3.1e-03 1.1e-02 2 3.2e-01 3.2e-03 1.1e-02 2 3.2e-01 3.2e-03 1.0e-02 1246974 1 3.5e-01 2.3e-03 1.2e-02 1 3.4e-01 2.2e-03 1.2e-02 1 3.5e-01 2.3e-03 1.3e-02 3234251 1 9.6e-01 6.1e-03 3.6e-02 1 9.3e-01 6.1e-03 3.2e-02 1 9.2e-01 6.0e-03 3.4e-02 8388608 1 2.2e+00 1.5e-02 7.1e-02 1 2.2e+00 1.5e-02 7.1e-02 1 2.2e+00 1.5e-02 7.5e-02 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 1.0e-02 4.0e-02 105 4.3e-01 3.5e-03 1.4e-02 109 4.5e-01 3.7e-03 1.4e-02 2 150 6.4e-01 5.2e-03 2.0e-02 74 3.1e-01 2.5e-03 9.6e-03 73 3.0e-01 2.5e-03 9.5e-03 4 75 3.2e-01 2.7e-03 1.0e-02 72 3.0e-01 2.5e-03 9.5e-03 71 3.0e-01 2.4e-03 9.3e-03 8 70 3.0e-01 2.5e-03 9.2e-03 72 3.0e-01 2.5e-03 9.4e-03 73 3.0e-01 2.5e-03 9.5e-03 16 70 3.1e-01 2.5e-03 9.6e-03 73 3.1e-01 2.5e-03 9.8e-03 73 3.1e-01 2.5e-03 9.8e-03 32 68 3.1e-01 2.5e-03 9.5e-03 72 3.1e-01 2.5e-03 9.8e-03 73 3.2e-01 2.5e-03 9.9e-03 64 67 3.0e-01 2.5e-03 9.3e-03 72 3.1e-01 2.5e-03 9.9e-03 71 3.1e-01 2.5e-03 9.7e-03 128 67 3.1e-01 2.5e-03 9.8e-03 71 3.2e-01 2.6e-03 1.0e-02 70 3.1e-01 2.6e-03 9.8e-03 256 65 3.1e-01 2.5e-03 9.7e-03 67 3.0e-01 2.4e-03 9.6e-03 68 3.2e-01 2.5e-03 1.0e-02 512 64 3.1e-01 2.5e-03 9.8e-03 68 3.1e-01 2.5e-03 9.9e-03 66 3.1e-01 2.5e-03 9.7e-03 1024 63 3.1e-01 2.5e-03 9.8e-03 68 3.2e-01 2.6e-03 1.0e-02 65 3.0e-01 2.5e-03 9.6e-03 2048 62 3.8e-01 2.7e-03 1.2e-02 64 3.9e-01 2.7e-03 1.2e-02 66 4.0e-01 2.8e-03 1.3e-02 4096 57 4.2e-01 2.9e-03 1.3e-02 58 4.3e-01 2.9e-03 1.4e-02 59 4.4e-01 2.9e-03 1.4e-02 10624 38 3.8e-01 2.6e-03 1.2e-02 39 3.8e-01 2.8e-03 1.3e-02 39 3.9e-01 2.8e-03 1.2e-02 27554 27 4.1e-01 2.6e-03 1.4e-02 27 4.1e-01 2.6e-03 1.3e-02 26 4.0e-01 2.7e-03 1.2e-02 71468 20 6.3e-01 3.6e-03 2.1e-02 19 5.8e-01 3.7e-03 1.8e-02 18 5.5e-01 3.2e-03 1.7e-02 185364 10 6.1e-01 4.2e-03 2.0e-02 9 5.3e-01 3.9e-03 1.7e-02 10 5.9e-01 4.3e-03 1.8e-02 480774 4 5.5e-01 3.9e-03 1.9e-02 4 5.3e-01 3.9e-03 1.8e-02 4 5.3e-01 3.9e-03 1.7e-02 1246974 1 3.2e-01 2.1e-03 1.3e-02 1 3.1e-01 2.1e-03 1.3e-02 1 3.1e-01 2.1e-03 1.3e-02 3234251 1 7.6e-01 6.1e-03 2.7e-02 1 7.6e-01 6.1e-03 2.7e-02 1 7.4e-01 6.2e-03 2.7e-02 8388608 1 1.9e+00 1.6e-02 6.4e-02 1 1.9e+00 1.6e-02 5.8e-02 1 1.9e+00 1.6e-02 6.3e-02 SECTION-LOOPLNGS-END SECTION-ELAPSED-BEGIN measurment loop: elapsed time = 98.622 sec sum of max elapsed time per entries above = 97.936 sec difference to elapsed time = 0.687 sec = 0.7% sum based on fastest repetition = 90.358 sec difference to elapsed time = 8.264 sec = 8.4% 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 -1 ) p02 ring-1*6fix 2 12 2.00 1.00 0 ( -1 -1 -1 ) p03 ring-1*6fix 2 12 2.00 1.00 0 ( -1 -1 -1 ) p04 ring-1*6fix 2 12 2.00 1.00 0 ( -1 -1 -1 ) p05 ring-1*6fix 2 12 2.00 1.00 0 ( -1 -1 -1 ) p06 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) 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 -1 ) p09 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) 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 -1 ) p15 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p16 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p17 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p18 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p19 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p20 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p21 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p22 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p23 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) 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 -1 ) p26 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) 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 -1 ) p30 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p31 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) 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 -1 ) p34 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p35 random-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p36 worst-cyc-1dim 2 12 2.00 1.00 0 ( -1 -1 -1 ) p37 best bi-section 2 6 1.00 0.50 0 ( -1 -1 -1 ) p38 worst bi-section 2 6 1.00 0.50 0 ( -1 -1 -1 ) p39 one PingPong Pair 2 2 1.00 0.50 4 ( -1 -1 -1 ) p40 acyclic-2dim-all 4 14 2.33 0.58 0 ( -1 -1 -1 ) p41 acyclic-3dim-all 4 14 2.33 0.58 0 ( -1 -1 -1 ) p42 cyclic-2dim-x 2 12 2.00 1.00 0 ( -1 -1 -1 ) p43 cyclic-2dim-y 1 6 1.00 1.00 0 ( -1 -1 -1 ) p44 cyclic-2dim-all 3 18 3.00 1.00 0 ( -1 -1 -1 ) p45 cyclic-3dim-x 2 12 2.00 1.00 0 ( -1 -1 -1 ) p46 cyclic-3dim-y 1 6 1.00 1.00 0 ( -1 -1 -1 ) p47 cyclic-3dim-all 3 18 3.00 1.00 0 ( -1 -1 -1 ) 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 : 145.421 80.486 129.100 -> 145.421 -> 872.529 MByte/s p01 ring-1*6fix : 134.940 94.934 127.814 -> 134.940 -> 809.643 MByte/s p02 ring-1*6fix : 138.000 96.247 132.410 -> 138.000 -> 828.000 MByte/s p03 ring-1*6fix : 135.381 94.555 131.461 -> 135.381 -> 812.287 MByte/s p04 ring-1*6fix : 138.164 95.064 131.175 -> 138.164 -> 828.982 MByte/s p05 ring-1*6fix : 136.254 94.945 132.889 -> 136.254 -> 817.525 MByte/s p06 random-cyc-1dim : 136.908 95.200 133.844 -> 136.908 -> 821.449 MByte/s p07 random-cyc-1dim : 134.517 93.494 129.866 -> 134.517 -> 807.104 MByte/s p08 random-cyc-1dim : 101.817 87.930 102.954 -> 102.954 -> 617.726 MByte/s p09 random-cyc-1dim : 105.781 87.618 99.361 -> 105.781 -> 634.686 MByte/s p10 random-cyc-1dim : 102.985 88.957 95.908 -> 102.985 -> 617.908 MByte/s p11 random-cyc-1dim : 104.988 88.813 99.872 -> 104.988 -> 629.929 MByte/s p12 random-cyc-1dim : 137.182 95.377 129.940 -> 137.182 -> 823.093 MByte/s p13 random-cyc-1dim : 102.034 90.228 104.531 -> 104.531 -> 627.189 MByte/s p14 random-cyc-1dim : 82.931 79.739 78.766 -> 82.931 -> 497.585 MByte/s p15 random-cyc-1dim : 103.115 88.267 99.956 -> 103.115 -> 618.692 MByte/s p16 random-cyc-1dim : 103.713 88.373 102.799 -> 103.713 -> 622.280 MByte/s p17 random-cyc-1dim : 136.683 95.131 131.531 -> 136.683 -> 820.097 MByte/s p18 random-cyc-1dim : 135.986 94.651 130.748 -> 135.986 -> 815.917 MByte/s p19 random-cyc-1dim : 103.820 89.318 99.086 -> 103.820 -> 622.923 MByte/s p20 random-cyc-1dim : 103.907 89.036 99.322 -> 103.907 -> 623.443 MByte/s p21 random-cyc-1dim : 105.128 88.574 98.325 -> 105.128 -> 630.768 MByte/s p22 random-cyc-1dim : 136.967 94.414 127.402 -> 136.967 -> 821.799 MByte/s p23 random-cyc-1dim : 103.476 88.016 103.671 -> 103.671 -> 622.029 MByte/s p24 random-cyc-1dim : 102.350 88.321 100.508 -> 102.350 -> 614.102 MByte/s p25 random-cyc-1dim : 135.978 96.056 132.541 -> 135.978 -> 815.869 MByte/s p26 random-cyc-1dim : 138.317 95.914 131.132 -> 138.317 -> 829.901 MByte/s p27 random-cyc-1dim : 136.322 94.430 129.154 -> 136.322 -> 817.933 MByte/s p28 random-cyc-1dim : 101.601 89.151 102.874 -> 102.874 -> 617.244 MByte/s p29 random-cyc-1dim : 135.144 94.461 125.755 -> 135.144 -> 810.866 MByte/s p30 random-cyc-1dim : 136.463 95.141 128.780 -> 136.463 -> 818.778 MByte/s p31 random-cyc-1dim : 137.883 96.451 129.469 -> 137.883 -> 827.296 MByte/s p32 random-cyc-1dim : 103.498 89.529 104.283 -> 104.283 -> 625.696 MByte/s p33 random-cyc-1dim : 101.261 89.608 99.397 -> 101.261 -> 607.567 MByte/s p34 random-cyc-1dim : 104.678 89.814 104.304 -> 104.678 -> 628.068 MByte/s p35 random-cyc-1dim : 84.008 78.601 88.584 -> 88.584 -> 531.507 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 83.091 79.323 81.331 -> 83.091 -> 498.547 MByte/s p37 best bi-section : 101.345 79.581 136.654 -> 136.654 -> 819.923 MByte/s p38 worst bi-section : 58.851 73.063 87.322 -> 87.322 -> 523.932 MByte/s p39 one PingPong Pair : 54.907 0.000 0.000 -> 54.907 -> 329.440 MByte/s p40 acyclic-2dim-all : 89.439 85.299 106.780 -> 106.780 -> 640.683 MByte/s p41 acyclic-3dim-all : 89.619 84.955 106.380 -> 106.380 -> 638.283 MByte/s p42 cyclic-2dim-x : 108.703 79.932 108.536 -> 108.703 -> 652.217 MByte/s p43 cyclic-2dim-y : 146.064 79.981 136.550 -> 146.064 -> 876.384 MByte/s p44 cyclic-2dim-all : 116.194 88.627 115.502 -> 116.194 -> 697.161 MByte/s p45 cyclic-3dim-x : 110.095 80.132 105.466 -> 110.095 -> 660.569 MByte/s p46 cyclic-3dim-y : 145.639 80.290 132.309 -> 145.639 -> 873.835 MByte/s p47 cyclic-3dim-all : 116.805 88.834 114.758 -> 116.805 -> 700.832 MByte/s log_avg of all rings : 137.983 92.530 130.796 || 137.983 -> 827.897 MByte/s log_avg of all random : 113.906 90.580 110.347 || 114.325 -> 685.949 MByte/s log_avg(ring,random) : 125.368 91.550 120.137 ||(125.598 -> 753.588)MByte/s * size -> accumulated on all pr.: 752.205 549.299 720.824 ||(753.588)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 : 133.812 145.264 145.774 -> 145.774 -> 874.644 MByte/s p01 ring-1*6fix : 125.683 132.929 133.968 -> 133.968 -> 803.807 MByte/s p02 ring-1*6fix : 134.239 136.755 135.205 -> 136.755 -> 820.530 MByte/s p03 ring-1*6fix : 135.620 134.716 133.981 -> 135.620 -> 813.720 MByte/s p04 ring-1*6fix : 137.809 135.414 135.422 -> 137.809 -> 826.856 MByte/s p05 ring-1*6fix : 129.863 137.360 133.337 -> 137.360 -> 824.163 MByte/s p06 random-cyc-1dim : 137.537 134.712 134.305 -> 137.537 -> 825.220 MByte/s p07 random-cyc-1dim : 121.209 136.166 133.281 -> 136.166 -> 816.995 MByte/s p08 random-cyc-1dim : 97.382 105.031 104.820 -> 105.031 -> 630.189 MByte/s p09 random-cyc-1dim : 104.290 105.175 106.652 -> 106.652 -> 639.912 MByte/s p10 random-cyc-1dim : 94.266 104.259 105.616 -> 105.616 -> 633.693 MByte/s p11 random-cyc-1dim : 101.979 107.161 106.514 -> 107.161 -> 642.966 MByte/s p12 random-cyc-1dim : 134.818 135.815 134.588 -> 135.815 -> 814.888 MByte/s p13 random-cyc-1dim : 103.354 107.812 106.693 -> 107.812 -> 646.873 MByte/s p14 random-cyc-1dim : 87.240 87.058 87.378 -> 87.378 -> 524.267 MByte/s p15 random-cyc-1dim : 103.080 104.363 104.958 -> 104.958 -> 629.750 MByte/s p16 random-cyc-1dim : 106.975 106.282 106.111 -> 106.975 -> 641.853 MByte/s p17 random-cyc-1dim : 133.358 136.097 133.327 -> 136.097 -> 816.581 MByte/s p18 random-cyc-1dim : 135.644 134.716 135.087 -> 135.644 -> 813.866 MByte/s p19 random-cyc-1dim : 105.278 104.990 106.331 -> 106.331 -> 637.986 MByte/s p20 random-cyc-1dim : 102.947 104.497 106.606 -> 106.606 -> 639.638 MByte/s p21 random-cyc-1dim : 104.454 105.624 105.757 -> 105.757 -> 634.541 MByte/s p22 random-cyc-1dim : 134.700 133.312 134.074 -> 134.700 -> 808.199 MByte/s p23 random-cyc-1dim : 105.444 105.520 104.891 -> 105.520 -> 633.120 MByte/s p24 random-cyc-1dim : 105.363 105.214 106.875 -> 106.875 -> 641.253 MByte/s p25 random-cyc-1dim : 135.496 136.129 135.238 -> 136.129 -> 816.776 MByte/s p26 random-cyc-1dim : 136.386 137.251 134.873 -> 137.251 -> 823.506 MByte/s p27 random-cyc-1dim : 135.516 135.812 135.035 -> 135.812 -> 814.869 MByte/s p28 random-cyc-1dim : 103.984 105.273 105.013 -> 105.273 -> 631.637 MByte/s p29 random-cyc-1dim : 130.265 134.192 133.883 -> 134.192 -> 805.151 MByte/s p30 random-cyc-1dim : 131.932 135.925 134.296 -> 135.925 -> 815.547 MByte/s p31 random-cyc-1dim : 136.523 137.034 136.261 -> 137.034 -> 822.205 MByte/s p32 random-cyc-1dim : 107.242 106.751 105.510 -> 107.242 -> 643.453 MByte/s p33 random-cyc-1dim : 104.674 105.961 103.635 -> 105.961 -> 635.764 MByte/s p34 random-cyc-1dim : 106.373 104.009 107.568 -> 107.568 -> 645.407 MByte/s p35 random-cyc-1dim : 88.613 90.971 89.309 -> 90.971 -> 545.825 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 87.337 87.204 86.769 -> 87.337 -> 524.022 MByte/s p37 best bi-section : 133.539 130.592 135.515 -> 135.515 -> 813.090 MByte/s p38 worst bi-section : 87.002 84.648 85.253 -> 87.002 -> 522.010 MByte/s p39 one PingPong Pair : 54.539 54.405 54.073 -> 54.539 -> 327.235 MByte/s p40 acyclic-2dim-all : 104.063 105.945 105.932 -> 105.945 -> 635.668 MByte/s p41 acyclic-3dim-all : 105.412 100.754 103.073 -> 105.412 -> 632.472 MByte/s p42 cyclic-2dim-x : 112.368 110.755 110.509 -> 112.368 -> 674.206 MByte/s p43 cyclic-2dim-y : 145.820 145.038 144.967 -> 145.820 -> 874.922 MByte/s p44 cyclic-2dim-all : 115.012 118.435 116.665 -> 118.435 -> 710.609 MByte/s p45 cyclic-3dim-x : 110.224 111.243 111.341 -> 111.341 -> 668.045 MByte/s p46 cyclic-3dim-y : 145.003 142.407 140.666 -> 145.003 -> 870.015 MByte/s p47 cyclic-3dim-all : 116.879 117.521 118.608 -> 118.608 -> 711.645 MByte/s log_avg of all rings : 132.777 137.018 136.215 || 137.831 -> 826.988 MByte/s log_avg of all random : 113.377 115.318 115.100 || 115.968 -> 695.806 MByte/s log_avg(ring,random) : 122.694 125.701 125.213 ||(126.428 -> 758.566)MByte/s * size -> accumulated on all pr.: 736.164 754.204 751.279 ||(758.566)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 :( 27.000) 0.037 0.589 8.520 61.571 416.550 494.962 -> 145.421 -> 872.529 MByte/s p00 method 1 =Alltoal :(144.605) 0.007 0.110 1.697 22.864 253.321 262.644 -> 80.486 -> 482.917 MByte/s p00 method 2 =non-blk :( 50.381) 0.020 0.296 4.774 46.361 319.283 501.441 -> 129.100 -> 774.600 MByte/s p01 ring-1*6fix p01 method 0 =Sndrcv :( 24.402) 0.041 0.606 8.986 72.217 307.567 492.418 -> 134.940 -> 809.643 MByte/s p01 method 1 =Alltoal :( 72.822) 0.014 0.218 3.378 43.344 256.692 330.540 -> 94.934 -> 569.601 MByte/s p01 method 2 =non-blk :( 43.082) 0.023 0.357 5.365 55.351 333.155 489.517 -> 127.814 -> 766.886 MByte/s p02 ring-1*6fix p02 method 0 =Sndrcv :( 24.373) 0.041 0.612 8.989 73.175 330.175 499.574 -> 138.000 -> 828.000 MByte/s p02 method 1 =Alltoal :( 73.588) 0.014 0.218 3.390 43.442 257.361 327.175 -> 96.247 -> 577.482 MByte/s p02 method 2 =non-blk :( 43.179) 0.023 0.358 5.368 55.410 313.431 486.946 -> 132.410 -> 794.459 MByte/s p03 ring-1*6fix p03 method 0 =Sndrcv :( 24.418) 0.041 0.610 8.980 74.171 302.051 495.474 -> 135.381 -> 812.287 MByte/s p03 method 1 =Alltoal :( 72.767) 0.014 0.214 3.342 43.657 255.937 322.614 -> 94.555 -> 567.328 MByte/s p03 method 2 =non-blk :( 43.018) 0.023 0.360 5.440 55.345 306.132 470.595 -> 131.461 -> 788.765 MByte/s p04 ring-1*6fix p04 method 0 =Sndrcv :( 24.374) 0.041 0.609 9.030 73.667 337.381 504.168 -> 138.164 -> 828.982 MByte/s p04 method 1 =Alltoal :( 72.839) 0.014 0.216 3.215 43.131 259.163 324.706 -> 95.064 -> 570.384 MByte/s p04 method 2 =non-blk :( 43.211) 0.023 0.359 5.442 55.761 325.769 479.774 -> 131.175 -> 787.050 MByte/s p05 ring-1*6fix p05 method 0 =Sndrcv :( 24.375) 0.041 0.609 8.948 73.776 309.845 505.049 -> 136.254 -> 817.525 MByte/s p05 method 1 =Alltoal :( 72.984) 0.014 0.216 3.381 43.166 263.863 318.021 -> 94.945 -> 569.672 MByte/s p05 method 2 =non-blk :( 43.206) 0.023 0.358 5.368 55.915 336.585 465.647 -> 132.889 -> 797.337 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 24.369) 0.041 0.609 9.015 74.250 321.255 489.431 -> 136.908 -> 821.449 MByte/s p06 method 1 =Alltoal :( 73.375) 0.014 0.217 3.383 43.639 260.117 330.644 -> 95.200 -> 571.198 MByte/s p06 method 2 =non-blk :( 43.207) 0.023 0.355 5.370 55.628 330.679 468.624 -> 133.844 -> 803.064 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 24.675) 0.041 0.603 8.954 72.577 309.243 503.382 -> 134.517 -> 807.104 MByte/s p07 method 1 =Alltoal :( 72.822) 0.014 0.218 3.375 42.604 238.795 318.528 -> 93.494 -> 560.962 MByte/s p07 method 2 =non-blk :( 43.481) 0.023 0.353 5.358 55.431 330.449 453.328 -> 129.866 -> 779.197 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 27.170) 0.037 0.566 8.234 61.580 262.092 366.019 -> 101.817 -> 610.904 MByte/s p08 method 1 =Alltoal :( 67.304) 0.015 0.234 3.590 37.321 243.622 310.270 -> 87.930 -> 527.579 MByte/s p08 method 2 =non-blk :( 46.500) 0.022 0.328 4.976 47.362 264.449 363.324 -> 102.954 -> 617.726 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 27.141) 0.037 0.568 8.360 61.090 260.109 385.727 -> 105.781 -> 634.686 MByte/s p09 method 1 =Alltoal :( 67.445) 0.015 0.234 3.558 37.285 237.116 303.539 -> 87.618 -> 525.708 MByte/s p09 method 2 =non-blk :( 46.605) 0.021 0.329 5.032 46.532 256.772 371.219 -> 99.361 -> 596.165 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 27.107) 0.037 0.569 8.348 60.804 247.469 389.380 -> 102.985 -> 617.908 MByte/s p10 method 1 =Alltoal :( 67.287) 0.015 0.235 3.577 36.854 247.608 307.619 -> 88.957 -> 533.739 MByte/s p10 method 2 =non-blk :( 46.191) 0.022 0.330 5.038 48.126 267.732 344.424 -> 95.908 -> 575.451 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 27.184) 0.037 0.568 8.316 60.703 270.639 389.561 -> 104.988 -> 629.929 MByte/s p11 method 1 =Alltoal :( 67.249) 0.015 0.235 3.542 36.712 240.146 319.743 -> 88.813 -> 532.879 MByte/s p11 method 2 =non-blk :( 46.583) 0.021 0.329 5.033 46.421 266.461 357.868 -> 99.872 -> 599.232 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 24.385) 0.041 0.610 8.976 74.755 339.862 491.625 -> 137.182 -> 823.093 MByte/s p12 method 1 =Alltoal :( 74.321) 0.013 0.215 3.372 43.525 263.349 319.074 -> 95.377 -> 572.265 MByte/s p12 method 2 =non-blk :( 43.060) 0.023 0.357 5.378 55.748 333.058 439.160 -> 129.940 -> 779.642 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 27.103) 0.037 0.572 8.356 61.245 205.847 389.425 -> 102.034 -> 612.205 MByte/s p13 method 1 =Alltoal :( 66.928) 0.015 0.230 3.600 36.747 247.067 320.954 -> 90.228 -> 541.369 MByte/s p13 method 2 =non-blk :( 46.807) 0.021 0.331 4.960 47.550 270.342 376.086 -> 104.531 -> 627.189 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 27.774) 0.036 0.553 8.061 60.936 192.223 302.762 -> 82.931 -> 497.585 MByte/s p14 method 1 =Alltoal :( 63.375) 0.016 0.242 3.791 34.534 227.894 275.524 -> 79.739 -> 478.431 MByte/s p14 method 2 =non-blk :( 48.138) 0.021 0.318 4.770 47.628 221.124 293.873 -> 78.766 -> 472.593 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 27.158) 0.037 0.568 8.357 60.900 235.147 388.721 -> 103.115 -> 618.692 MByte/s p15 method 1 =Alltoal :( 67.179) 0.015 0.227 3.600 37.043 244.018 308.184 -> 88.267 -> 529.602 MByte/s p15 method 2 =non-blk :( 46.926) 0.021 0.333 4.952 47.273 258.488 362.195 -> 99.956 -> 599.735 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 27.080) 0.037 0.567 8.317 60.468 227.162 385.701 -> 103.713 -> 622.280 MByte/s p16 method 1 =Alltoal :( 68.137) 0.015 0.227 3.579 37.151 238.142 307.659 -> 88.373 -> 530.238 MByte/s p16 method 2 =non-blk :( 46.872) 0.021 0.332 4.957 47.209 274.796 380.860 -> 102.799 -> 616.793 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 24.401) 0.041 0.607 9.074 74.296 328.287 492.794 -> 136.683 -> 820.097 MByte/s p17 method 1 =Alltoal :( 72.786) 0.014 0.214 3.333 43.212 257.761 326.202 -> 95.131 -> 570.785 MByte/s p17 method 2 =non-blk :( 43.587) 0.023 0.356 5.455 55.663 331.481 485.536 -> 131.531 -> 789.187 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 24.592) 0.041 0.605 9.009 73.442 298.997 503.457 -> 135.986 -> 815.917 MByte/s p18 method 1 =Alltoal :( 73.107) 0.014 0.217 3.325 42.713 258.388 329.689 -> 94.651 -> 567.909 MByte/s p18 method 2 =non-blk :( 43.667) 0.023 0.353 5.429 54.895 334.411 472.079 -> 130.748 -> 784.488 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 27.076) 0.037 0.568 8.361 61.938 247.550 382.928 -> 103.820 -> 622.923 MByte/s p19 method 1 =Alltoal :( 68.555) 0.015 0.235 3.620 37.213 232.507 319.779 -> 89.318 -> 535.906 MByte/s p19 method 2 =non-blk :( 46.509) 0.022 0.330 4.979 46.564 256.284 365.628 -> 99.086 -> 594.514 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 27.101) 0.037 0.569 8.385 61.148 246.972 383.480 -> 103.907 -> 623.443 MByte/s p20 method 1 =Alltoal :( 67.089) 0.015 0.234 3.607 37.432 241.592 319.840 -> 89.036 -> 534.217 MByte/s p20 method 2 =non-blk :( 46.404) 0.022 0.330 4.985 47.009 266.635 370.882 -> 99.322 -> 595.933 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 27.113) 0.037 0.569 8.341 61.214 260.809 386.082 -> 105.128 -> 630.768 MByte/s p21 method 1 =Alltoal :( 67.125) 0.015 0.234 3.607 37.516 231.236 319.627 -> 88.574 -> 531.446 MByte/s p21 method 2 =non-blk :( 46.601) 0.021 0.331 4.972 46.905 267.096 368.778 -> 98.325 -> 589.950 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 24.467) 0.041 0.605 8.993 74.414 328.414 498.357 -> 136.967 -> 821.799 MByte/s p22 method 1 =Alltoal :( 73.965) 0.014 0.217 3.357 43.657 259.255 319.043 -> 94.414 -> 566.483 MByte/s p22 method 2 =non-blk :( 43.170) 0.023 0.355 5.392 55.715 310.102 458.744 -> 127.402 -> 764.415 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 27.082) 0.037 0.568 8.315 61.130 244.515 382.343 -> 103.476 -> 620.856 MByte/s p23 method 1 =Alltoal :( 67.055) 0.015 0.235 3.608 37.358 239.296 309.748 -> 88.016 -> 528.096 MByte/s p23 method 2 =non-blk :( 46.211) 0.022 0.330 4.989 47.555 272.705 365.755 -> 103.671 -> 622.029 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 27.098) 0.037 0.568 8.304 61.589 221.319 386.233 -> 102.350 -> 614.102 MByte/s p24 method 1 =Alltoal :( 67.160) 0.015 0.234 3.514 36.771 226.051 306.092 -> 88.321 -> 529.927 MByte/s p24 method 2 =non-blk :( 46.450) 0.022 0.329 4.993 48.188 268.801 362.178 -> 100.508 -> 603.046 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 24.500) 0.041 0.609 9.059 73.679 315.001 503.987 -> 135.978 -> 815.869 MByte/s p25 method 1 =Alltoal :( 72.716) 0.014 0.216 3.238 43.181 269.377 316.910 -> 96.056 -> 576.337 MByte/s p25 method 2 =non-blk :( 43.101) 0.023 0.356 5.449 54.883 311.796 476.748 -> 132.541 -> 795.249 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 24.338) 0.041 0.614 8.985 73.660 344.951 494.350 -> 138.317 -> 829.901 MByte/s p26 method 1 =Alltoal :( 72.716) 0.014 0.213 3.373 43.639 264.570 329.876 -> 95.914 -> 575.482 MByte/s p26 method 2 =non-blk :( 43.362) 0.023 0.359 5.397 55.988 321.061 443.162 -> 131.132 -> 786.794 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 24.381) 0.041 0.606 8.955 73.970 309.173 503.684 -> 136.322 -> 817.933 MByte/s p27 method 1 =Alltoal :( 73.035) 0.014 0.217 3.366 43.608 261.957 317.931 -> 94.430 -> 566.577 MByte/s p27 method 2 =non-blk :( 43.091) 0.023 0.359 5.381 55.788 320.228 466.616 -> 129.154 -> 774.922 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 27.084) 0.037 0.568 8.317 61.265 232.285 388.830 -> 101.601 -> 609.604 MByte/s p28 method 1 =Alltoal :( 67.179) 0.015 0.229 3.592 36.853 244.705 311.896 -> 89.151 -> 534.906 MByte/s p28 method 2 =non-blk :( 46.890) 0.021 0.330 4.918 47.052 276.801 368.593 -> 102.874 -> 617.244 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 24.451) 0.041 0.606 8.998 73.841 316.434 489.859 -> 135.144 -> 810.866 MByte/s p29 method 1 =Alltoal :( 72.892) 0.014 0.217 3.376 43.625 257.181 320.899 -> 94.461 -> 566.766 MByte/s p29 method 2 =non-blk :( 43.266) 0.023 0.357 5.450 55.650 309.997 444.194 -> 125.755 -> 754.531 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 24.529) 0.041 0.607 9.042 73.628 308.752 503.110 -> 136.463 -> 818.778 MByte/s p30 method 1 =Alltoal :( 72.713) 0.014 0.218 3.316 43.098 258.393 318.366 -> 95.141 -> 570.843 MByte/s p30 method 2 =non-blk :( 43.450) 0.023 0.353 5.442 54.993 325.231 476.016 -> 128.780 -> 772.680 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 24.488) 0.041 0.610 9.036 74.419 334.745 503.157 -> 137.883 -> 827.296 MByte/s p31 method 1 =Alltoal :( 72.928) 0.014 0.217 3.306 43.624 269.617 328.907 -> 96.451 -> 578.708 MByte/s p31 method 2 =non-blk :( 43.771) 0.023 0.353 5.413 55.780 331.797 463.126 -> 129.469 -> 776.812 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 27.082) 0.037 0.567 8.364 60.935 240.251 375.245 -> 103.498 -> 620.988 MByte/s p32 method 1 =Alltoal :( 67.053) 0.015 0.235 3.601 36.984 236.890 320.213 -> 89.529 -> 537.177 MByte/s p32 method 2 =non-blk :( 46.849) 0.021 0.328 4.915 47.633 277.515 366.419 -> 104.283 -> 625.696 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 27.126) 0.037 0.566 8.249 61.420 218.894 377.152 -> 101.261 -> 607.567 MByte/s p33 method 1 =Alltoal :( 67.466) 0.015 0.234 3.600 37.081 237.224 312.158 -> 89.608 -> 537.651 MByte/s p33 method 2 =non-blk :( 46.457) 0.022 0.328 4.915 47.907 264.447 372.703 -> 99.397 -> 596.385 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 27.134) 0.037 0.567 8.296 61.267 252.837 386.662 -> 104.678 -> 628.068 MByte/s p34 method 1 =Alltoal :( 67.089) 0.015 0.235 3.611 36.478 242.029 319.773 -> 89.814 -> 538.886 MByte/s p34 method 2 =non-blk :( 46.500) 0.022 0.330 4.936 47.109 268.169 379.343 -> 104.304 -> 625.823 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 27.978) 0.036 0.550 8.110 60.870 203.789 301.169 -> 84.008 -> 504.050 MByte/s p35 method 1 =Alltoal :( 63.481) 0.016 0.248 3.792 33.732 221.893 280.461 -> 78.601 -> 471.607 MByte/s p35 method 2 =non-blk :( 48.091) 0.021 0.314 4.743 46.671 214.852 312.059 -> 88.584 -> 531.507 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 27.736) 0.036 0.551 8.084 61.388 196.398 302.260 -> 83.091 -> 498.547 MByte/s p36 method 1 =Alltoal :( 63.430) 0.016 0.248 3.791 34.441 227.439 275.710 -> 79.323 -> 475.937 MByte/s p36 method 2 =non-blk :( 48.267) 0.021 0.314 4.774 47.276 221.502 298.103 -> 81.331 -> 487.986 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 23.437) 0.021 0.319 4.650 38.560 239.927 390.168 -> 101.345 -> 608.073 MByte/s p37 method 1 =Alltoal :( 72.733) 0.007 0.107 1.705 22.738 239.490 264.466 -> 79.581 -> 477.485 MByte/s p37 method 2 =non-blk :( 25.052) 0.020 0.304 4.798 47.666 367.062 502.763 -> 136.654 -> 819.923 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 24.245) 0.021 0.306 4.469 37.210 195.213 162.962 -> 58.851 -> 353.108 MByte/s p38 method 1 =Alltoal :( 73.540) 0.007 0.109 1.681 21.895 193.137 296.564 -> 73.063 -> 438.377 MByte/s p38 method 2 =non-blk :( 25.394) 0.020 0.294 4.585 47.377 260.600 307.050 -> 87.322 -> 523.932 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 11.555) 0.014 0.220 3.142 33.683 115.100 203.760 -> 54.907 -> 329.440 MByte/s p39 method 1 =Alltoal :(-------) 0.000 0.000 0.000 0.000 0.000 0.000 -> 0.000 -> 0.000 MByte/s p39 method 2 =non-blk :(-------) 0.000 0.000 0.000 0.000 0.000 0.000 -> 0.000 -> 0.000 MByte/s p40 acyclic-2dim-all p40 method 0 =Sndrcv :( 23.967) 0.024 0.362 5.355 44.901 218.012 321.624 -> 89.439 -> 536.637 MByte/s p40 method 1 =Alltoal :( 37.196) 0.016 0.249 3.757 35.633 233.037 299.626 -> 85.299 -> 511.795 MByte/s p40 method 2 =non-blk :( 29.495) 0.020 0.307 4.563 43.799 289.735 369.004 -> 106.780 -> 640.683 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 24.001) 0.024 0.362 5.382 45.026 221.151 322.531 -> 89.619 -> 537.715 MByte/s p41 method 1 =Alltoal :( 37.550) 0.016 0.249 3.710 35.483 226.123 298.849 -> 84.955 -> 509.730 MByte/s p41 method 2 =non-blk :( 29.571) 0.020 0.307 4.554 43.852 280.617 371.540 -> 106.380 -> 638.283 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 27.412) 0.036 0.567 8.304 60.429 276.889 385.434 -> 108.703 -> 652.217 MByte/s p42 method 1 =Alltoal :( 74.247) 0.013 0.216 3.260 34.062 221.132 286.442 -> 79.932 -> 479.595 MByte/s p42 method 2 =non-blk :( 46.623) 0.021 0.327 4.988 47.628 265.907 402.631 -> 108.536 -> 651.213 MByte/s p43 cyclic-2dim-y p43 method 0 =Sndrcv :( 26.582) 0.038 0.590 8.637 61.401 405.321 498.995 -> 146.064 -> 876.384 MByte/s p43 method 1 =Alltoal :(146.074) 0.007 0.108 1.724 22.783 247.978 263.189 -> 79.981 -> 479.883 MByte/s p43 method 2 =non-blk :( 50.623) 0.020 0.301 4.657 45.971 395.822 501.052 -> 136.550 -> 819.302 MByte/s p44 cyclic-2dim-all p44 method 0 =Sndrcv :( 25.468) 0.039 0.592 8.783 67.498 284.776 409.593 -> 116.194 -> 697.161 MByte/s p44 method 1 =Alltoal :( 49.840) 0.020 0.315 4.807 45.034 233.777 292.551 -> 88.627 -> 531.762 MByte/s p44 method 2 =non-blk :( 43.156) 0.023 0.358 5.366 53.516 298.750 409.001 -> 115.502 -> 693.012 MByte/s p45 cyclic-3dim-x p45 method 0 =Sndrcv :( 27.539) 0.036 0.566 8.295 60.141 278.703 389.235 -> 110.095 -> 660.569 MByte/s p45 method 1 =Alltoal :( 73.054) 0.014 0.216 3.211 34.133 227.439 280.326 -> 80.132 -> 480.791 MByte/s p45 method 2 =non-blk :( 46.578) 0.021 0.328 5.051 48.109 271.597 382.186 -> 105.466 -> 632.795 MByte/s p46 cyclic-3dim-y p46 method 0 =Sndrcv :( 26.794) 0.037 0.590 8.430 61.715 401.540 501.681 -> 145.639 -> 873.835 MByte/s p46 method 1 =Alltoal :(144.465) 0.007 0.109 1.723 22.964 252.970 263.635 -> 80.290 -> 481.739 MByte/s p46 method 2 =non-blk :( 51.082) 0.020 0.296 4.753 46.835 366.090 499.530 -> 132.309 -> 793.853 MByte/s p47 cyclic-3dim-all p47 method 0 =Sndrcv :( 25.502) 0.039 0.594 8.678 67.617 288.456 413.537 -> 116.805 -> 700.832 MByte/s p47 method 1 =Alltoal :( 49.023) 0.020 0.316 4.845 45.165 247.124 290.180 -> 88.834 -> 533.005 MByte/s p47 method 2 =non-blk :( 43.156) 0.023 0.359 5.371 53.479 305.211 409.426 -> 114.758 -> 688.545 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.040 0.606 8.907 71.280 331.856 498.585 || 137.983 -> 827.897 MByte/s - ring, method 1 = Alltoal: 0.012 0.193 2.984 38.964 257.703 313.329 || 92.530 -> 555.181 MByte/s - ring, method 2 = non-blk: 0.023 0.347 5.287 53.906 322.215 482.171 || 130.796 -> 784.775 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.038 0.582 8.576 65.958 267.036 419.281 || 113.906 -> 683.433 MByte/s - random, method 1 = Alltoal: 0.014 0.227 3.501 39.233 246.247 314.724 || 90.580 -> 543.479 MByte/s - random, method 2 = non-blk: 0.022 0.339 5.127 50.401 284.787 397.392 || 110.347 -> 662.084 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.039 0.594 8.740 68.567 297.687 457.217 || 125.368 -> 752.205 MByte/s - average, method 1 = Alltoal: 0.013 0.209 3.232 39.098 251.910 314.026 || 91.550 -> 549.299 MByte/s - average, method 2 = non-blk: 0.022 0.343 5.207 52.124 302.924 437.734 || 120.137 -> 720.824 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 0.236 3.564 52.438 411.403 1786.123 2743.302 || 752.205 MByte/s - accumulated, mthd 1 = Alltoal: 0.080 1.256 19.391 234.590 1511.459 1884.155 || 549.299 MByte/s - accumulated, mthd 2 = non-blk: 0.134 2.058 31.239 312.743 1817.542 2626.404 || 720.824 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 0.236 0.039 0.040 0.038 0.039 0.013 0.022 2 0.469 0.078 0.080 0.076 0.078 0.026 0.044 4 0.930 0.155 0.159 0.151 0.155 0.053 0.088 8 1.885 0.314 0.323 0.306 0.314 0.106 0.178 16 3.564 0.594 0.606 0.582 0.594 0.209 0.343 32 7.042 1.174 1.194 1.154 1.174 0.418 0.679 64 13.827 2.304 2.349 2.261 2.304 0.832 1.350 128 26.504 4.417 4.497 4.339 4.417 1.648 2.638 256 52.438 8.740 8.907 8.576 8.740 3.232 5.207 512 102.986 17.164 17.524 16.812 17.164 6.446 10.237 1024 199.429 33.238 33.846 32.641 33.238 12.756 20.209 2048 252.724 42.121 43.593 40.698 42.121 21.664 31.126 4096 411.403 68.567 71.280 65.958 68.567 39.098 52.124 10624 619.883 103.314 109.951 97.077 99.847 69.806 100.195 27554 1050.792 175.132 187.111 163.920 160.690 122.543 170.752 71468 1390.368 231.728 254.214 211.231 223.465 188.023 227.409 185364 1880.553 313.425 341.741 287.456 297.687 251.910 302.924 480774 2234.676 372.446 410.181 338.182 369.976 287.264 350.525 1246974 2491.714 415.286 455.986 378.218 411.246 300.240 385.581 3234251 2596.771 432.795 475.043 394.304 430.535 300.005 418.311 8388608 2747.903 457.984 499.667 419.778 457.217 314.026 437.734 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 :( 27.000) 0.037 0.589 8.520 61.571 416.550 501.441 -> 146.797 -> 880.783 MByte/s p01 ring-1*6fix :( 24.402) 0.041 0.606 8.986 72.217 333.155 492.418 -> 136.584 -> 819.505 MByte/s p02 ring-1*6fix :( 24.373) 0.041 0.612 8.989 73.175 330.175 499.574 -> 138.177 -> 829.061 MByte/s p03 ring-1*6fix :( 24.418) 0.041 0.610 8.980 74.171 306.132 495.474 -> 136.243 -> 817.461 MByte/s p04 ring-1*6fix :( 24.374) 0.041 0.609 9.030 73.667 337.381 504.168 -> 138.799 -> 832.793 MByte/s p05 ring-1*6fix :( 24.375) 0.041 0.609 8.948 73.776 336.585 505.049 -> 138.014 -> 828.083 MByte/s p06 random-cyc-1dim :( 24.369) 0.041 0.609 9.015 74.250 330.679 489.431 -> 138.280 -> 829.681 MByte/s p07 random-cyc-1dim :( 24.675) 0.041 0.603 8.954 72.577 330.449 503.382 -> 136.531 -> 819.188 MByte/s p08 random-cyc-1dim :( 27.170) 0.037 0.566 8.234 61.580 264.449 366.019 -> 106.213 -> 637.277 MByte/s p09 random-cyc-1dim :( 27.141) 0.037 0.568 8.360 61.090 260.109 385.727 -> 107.618 -> 645.711 MByte/s p10 random-cyc-1dim :( 27.107) 0.037 0.569 8.348 60.804 267.732 389.380 -> 106.820 -> 640.921 MByte/s p11 random-cyc-1dim :( 27.184) 0.037 0.568 8.316 60.703 270.639 389.561 -> 108.074 -> 648.443 MByte/s p12 random-cyc-1dim :( 24.385) 0.041 0.610 8.976 74.755 339.862 491.625 -> 138.118 -> 828.708 MByte/s p13 random-cyc-1dim :( 27.103) 0.037 0.572 8.356 61.245 270.342 389.425 -> 109.180 -> 655.082 MByte/s p14 random-cyc-1dim :( 27.774) 0.036 0.553 8.061 60.936 227.894 302.762 -> 88.666 -> 531.996 MByte/s p15 random-cyc-1dim :( 27.158) 0.037 0.568 8.357 60.900 258.488 388.721 -> 107.231 -> 643.389 MByte/s p16 random-cyc-1dim :( 27.080) 0.037 0.567 8.317 60.468 274.796 385.701 -> 108.857 -> 653.144 MByte/s p17 random-cyc-1dim :( 24.401) 0.041 0.607 9.074 74.296 331.481 492.794 -> 137.481 -> 824.889 MByte/s p18 random-cyc-1dim :( 24.592) 0.041 0.605 9.009 73.442 334.411 503.457 -> 137.959 -> 827.756 MByte/s p19 random-cyc-1dim :( 27.076) 0.037 0.568 8.361 61.938 256.284 382.928 -> 107.274 -> 643.646 MByte/s p20 random-cyc-1dim :( 27.101) 0.037 0.569 8.385 61.148 266.635 383.480 -> 107.363 -> 644.176 MByte/s p21 random-cyc-1dim :( 27.113) 0.037 0.569 8.341 61.214 267.096 386.082 -> 108.099 -> 648.593 MByte/s p22 random-cyc-1dim :( 24.467) 0.041 0.605 8.993 74.414 328.414 498.357 -> 137.234 -> 823.407 MByte/s p23 random-cyc-1dim :( 27.082) 0.037 0.568 8.315 61.130 272.705 382.343 -> 108.160 -> 648.958 MByte/s p24 random-cyc-1dim :( 27.098) 0.037 0.568 8.304 61.589 268.801 386.233 -> 107.455 -> 644.733 MByte/s p25 random-cyc-1dim :( 24.500) 0.041 0.609 9.059 73.679 315.001 503.987 -> 137.357 -> 824.142 MByte/s p26 random-cyc-1dim :( 24.338) 0.041 0.614 8.985 73.660 344.951 494.350 -> 138.914 -> 833.484 MByte/s p27 random-cyc-1dim :( 24.381) 0.041 0.606 8.955 73.970 320.228 503.684 -> 137.411 -> 824.464 MByte/s p28 random-cyc-1dim :( 27.084) 0.037 0.568 8.317 61.265 276.801 388.830 -> 107.729 -> 646.374 MByte/s p29 random-cyc-1dim :( 24.451) 0.041 0.606 8.998 73.841 316.434 489.859 -> 135.465 -> 812.791 MByte/s p30 random-cyc-1dim :( 24.529) 0.041 0.607 9.042 73.628 325.231 503.110 -> 137.469 -> 824.816 MByte/s p31 random-cyc-1dim :( 24.488) 0.041 0.610 9.036 74.419 334.745 503.157 -> 138.201 -> 829.206 MByte/s p32 random-cyc-1dim :( 27.082) 0.037 0.567 8.364 60.935 277.515 375.245 -> 108.254 -> 649.526 MByte/s p33 random-cyc-1dim :( 27.126) 0.037 0.566 8.249 61.420 264.447 377.152 -> 107.219 -> 643.315 MByte/s p34 random-cyc-1dim :( 27.134) 0.037 0.567 8.296 61.267 268.169 386.662 -> 108.909 -> 653.454 MByte/s p35 random-cyc-1dim :( 27.978) 0.036 0.550 8.110 60.870 221.893 312.059 -> 91.282 -> 547.692 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 27.736) 0.036 0.551 8.084 61.388 227.439 302.260 -> 88.097 -> 528.580 MByte/s p37 best bi-section :( 23.437) 0.021 0.319 4.798 47.666 367.062 502.763 -> 136.658 -> 819.949 MByte/s p38 worst bi-section :( 24.245) 0.021 0.306 4.585 47.377 260.600 307.050 -> 88.565 -> 531.388 MByte/s p39 one PingPong Pair :( 11.555) 0.014 0.220 3.142 33.683 115.100 203.760 -> 54.907 -> 329.440 MByte/s p40 acyclic-2dim-all :( 23.967) 0.024 0.362 5.355 44.901 289.735 369.004 -> 107.139 -> 642.833 MByte/s p41 acyclic-3dim-all :( 24.001) 0.024 0.362 5.382 45.026 280.617 371.540 -> 106.734 -> 640.405 MByte/s p42 cyclic-2dim-x :( 27.412) 0.036 0.567 8.304 60.429 276.889 402.631 -> 114.165 -> 684.992 MByte/s p43 cyclic-2dim-y :( 26.582) 0.038 0.590 8.637 61.401 405.321 501.052 -> 146.410 -> 878.457 MByte/s p44 cyclic-2dim-all :( 25.468) 0.039 0.592 8.783 67.498 298.750 409.593 -> 119.057 -> 714.339 MByte/s p45 cyclic-3dim-x :( 27.539) 0.036 0.566 8.295 60.141 278.703 389.235 -> 112.513 -> 675.075 MByte/s p46 cyclic-3dim-y :( 26.794) 0.037 0.590 8.430 61.715 401.540 501.681 -> 145.897 -> 875.381 MByte/s p47 cyclic-3dim-all :( 25.502) 0.039 0.594 8.678 67.617 305.211 413.537 -> 119.416 -> 716.494 MByte/s log_avg of all rings : 0.040 0.606 8.907 71.280 341.741 499.667 || 139.058 -> 834.349 MByte/s log_avg of all random : 0.038 0.582 8.576 65.958 287.456 419.778 || 117.394 -> 704.362 MByte/s log_avg(ring,random) : 0.039 0.594 8.740 68.567 313.425 457.984 || 127.768 -> 766.605 MByte/s * size -> accumulated on all pr.: 0.236 3.564 52.438 411.403 1880.553 2747.903 || 766.605 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 766.605 MByte/s on 6 processes ( = 127.768 MByte/s * 6 processes) Ping-pong latency: 11.555 microsec Ping-pong bandwidth: 1222.560 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 17:28:28 1999 Total execution wall clock time = 100 seconds SECTION-BEFF-END b_eff = 766.605 MB/s = 127.768 * 6 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000