b_eff = 680.633 MB/s = 85.079 * 8 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 = 8 2-dim-paterns: size = 4 * 2 3-dim-paterns: size = 2 * 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-4*2fix 1=ring-2*4fix 2=ring-1*8fix 3=ring-1*8fix 4=ring-1*8fix 5=ring-1*8fix 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: 87.059 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 7.7e-01 4.7e-03 2.9e-02 238 6.2e-01 3.8e-03 2.4e-02 240 6.2e-01 3.7e-03 2.3e-02 2 160 4.1e-01 2.5e-03 1.6e-02 158 4.1e-01 2.5e-03 1.5e-02 161 4.2e-01 2.5e-03 1.6e-02 4 162 4.2e-01 2.5e-03 1.6e-02 160 4.2e-01 2.5e-03 1.6e-02 161 4.2e-01 2.5e-03 1.6e-02 8 161 4.1e-01 2.5e-03 1.6e-02 159 4.1e-01 2.5e-03 1.6e-02 160 4.1e-01 2.5e-03 1.6e-02 16 159 4.3e-01 2.5e-03 1.6e-02 160 4.3e-01 2.5e-03 1.7e-02 162 4.3e-01 2.5e-03 1.7e-02 32 160 4.3e-01 2.6e-03 1.7e-02 158 4.3e-01 2.6e-03 1.7e-02 158 4.3e-01 2.5e-03 1.6e-02 64 154 4.2e-01 2.5e-03 1.7e-02 151 4.2e-01 2.5e-03 1.7e-02 157 4.3e-01 2.6e-03 1.7e-02 128 151 4.3e-01 2.7e-03 1.8e-02 150 4.3e-01 2.6e-03 1.7e-02 151 4.3e-01 2.6e-03 1.8e-02 256 140 4.2e-01 2.5e-03 1.6e-02 142 4.2e-01 2.4e-03 1.6e-02 143 4.3e-01 2.5e-03 1.7e-02 512 142 4.3e-01 2.6e-03 1.7e-02 145 4.3e-01 2.7e-03 1.7e-02 142 4.4e-01 2.6e-03 1.7e-02 1024 138 4.3e-01 2.6e-03 1.7e-02 135 4.2e-01 2.6e-03 1.6e-02 138 4.4e-01 2.6e-03 1.7e-02 2048 131 6.5e-01 3.3e-03 2.2e-02 132 6.6e-01 3.3e-03 2.2e-02 131 6.5e-01 3.3e-03 2.2e-02 4096 100 6.2e-01 3.2e-03 2.1e-02 99 6.4e-01 3.3e-03 2.2e-02 100 6.4e-01 3.2e-03 2.2e-02 10624 59 6.8e-01 2.9e-03 2.2e-02 58 6.7e-01 2.9e-03 2.1e-02 59 6.8e-01 3.0e-03 2.2e-02 27554 38 7.6e-01 3.1e-03 2.4e-02 38 7.6e-01 3.1e-03 2.5e-02 38 7.5e-01 3.2e-03 2.4e-02 71468 23 9.1e-01 3.8e-03 3.0e-02 23 9.1e-01 3.6e-03 3.1e-02 22 8.5e-01 3.5e-03 2.9e-02 185364 11 8.7e-01 4.2e-03 2.9e-02 12 9.3e-01 5.1e-03 3.3e-02 12 9.2e-01 4.8e-03 3.1e-02 480774 5 8.7e-01 5.1e-03 3.2e-02 4 7.0e-01 3.8e-03 2.5e-02 4 6.9e-01 3.9e-03 2.5e-02 1246974 1 4.2e-01 2.4e-03 1.7e-02 2 8.2e-01 4.5e-03 3.4e-02 1 3.9e-01 2.4e-03 1.6e-02 3234251 1 4.8e-01 8.5e-03 3.5e-02 M 1 5.9e-01 8.6e-03 3.2e-02 M 1 5.9e-01 8.6e-03 3.5e-02 M 8388608 1 1.2e+00 2.3e-02 9.7e-02 R 1 1.4e+00 2.2e-02 7.3e-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=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 2.6e+00 5.1e-02 6.1e-02 27 2.4e-01 4.6e-03 5.5e-03 21 1.8e-01 3.6e-03 4.3e-03 2 150 1.3e+00 2.6e-02 3.1e-02 14 1.2e-01 2.4e-03 2.9e-03 14 1.2e-01 2.4e-03 3.0e-03 4 75 6.6e-01 1.3e-02 1.5e-02 14 1.2e-01 2.4e-03 2.9e-03 14 1.2e-01 2.4e-03 3.0e-03 8 37 3.2e-01 6.3e-03 7.5e-03 14 1.2e-01 2.4e-03 2.9e-03 14 1.2e-01 2.4e-03 2.9e-03 16 18 1.6e-01 3.1e-03 3.7e-03 14 1.3e-01 2.4e-03 3.1e-03 14 1.2e-01 2.4e-03 3.0e-03 32 14 1.2e-01 2.4e-03 2.9e-03 14 1.2e-01 2.4e-03 3.0e-03 14 1.2e-01 2.4e-03 2.9e-03 64 14 1.2e-01 2.4e-03 2.9e-03 14 1.3e-01 2.4e-03 3.1e-03 14 1.3e-01 2.4e-03 3.0e-03 128 14 1.3e-01 2.4e-03 3.0e-03 14 1.3e-01 2.4e-03 2.9e-03 14 1.3e-01 2.4e-03 3.0e-03 256 14 1.3e-01 2.4e-03 3.0e-03 14 1.3e-01 2.4e-03 3.0e-03 14 1.3e-01 2.5e-03 3.1e-03 512 14 1.3e-01 2.5e-03 2.9e-03 14 1.3e-01 2.5e-03 3.0e-03 14 1.3e-01 2.5e-03 3.0e-03 1024 14 1.3e-01 2.5e-03 3.1e-03 14 1.3e-01 2.5e-03 3.1e-03 14 1.3e-01 2.5e-03 3.1e-03 2048 14 1.7e-01 2.7e-03 3.9e-03 14 1.7e-01 2.7e-03 4.0e-03 14 1.7e-01 2.7e-03 4.1e-03 4096 13 1.8e-01 2.8e-03 4.3e-03 13 1.8e-01 2.7e-03 4.4e-03 13 1.8e-01 2.7e-03 4.4e-03 10624 8 1.5e-01 2.1e-03 4.6e-03 9 1.8e-01 2.4e-03 8.1e-03 9 1.7e-01 2.3e-03 4.2e-03 27554 7 2.0e-01 2.2e-03 5.8e-03 7 2.0e-01 2.2e-03 5.1e-03 7 2.0e-01 2.2e-03 5.2e-03 71468 6 3.0e-01 3.2e-03 9.3e-03 6 3.0e-01 2.9e-03 8.2e-03 6 2.9e-01 3.1e-03 7.8e-03 185364 3 2.9e-01 2.7e-03 8.9e-03 3 2.9e-01 2.6e-03 9.4e-03 3 2.8e-01 2.9e-03 7.9e-03 480774 2 4.2e-01 3.7e-03 1.2e-02 2 4.2e-01 4.0e-03 1.3e-02 2 4.2e-01 3.9e-03 1.3e-02 1246974 1 4.9e-01 4.5e-03 1.7e-02 1 5.0e-01 4.5e-03 1.8e-02 1 4.9e-01 4.4e-03 1.7e-02 3234251 1 1.6e-01 2.1e-02 3.6e-02 M 1 8.9e-02 2.0e-02 2.3e-02 M 1 1.4e-01 2.0e-02 2.8e-02 M 8388608 1 3.6e-01 5.1e-02 7.5e-02 R 1 2.2e-01 5.2e-02 5.7e-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 1.3e+00 1.1e-02 3.7e-02 103 4.5e-01 3.7e-03 1.3e-02 103 4.5e-01 3.6e-03 1.3e-02 2 150 6.6e-01 5.5e-03 1.9e-02 69 3.0e-01 2.5e-03 8.8e-03 70 3.1e-01 2.5e-03 8.7e-03 4 75 3.3e-01 2.8e-03 9.4e-03 70 3.1e-01 2.6e-03 8.9e-03 69 3.0e-01 2.5e-03 8.6e-03 8 67 2.9e-01 2.4e-03 8.3e-03 68 3.0e-01 2.4e-03 8.5e-03 70 3.0e-01 2.5e-03 8.6e-03 16 69 3.1e-01 2.5e-03 8.8e-03 69 3.1e-01 2.5e-03 8.8e-03 70 3.1e-01 2.5e-03 8.9e-03 32 67 3.1e-01 2.5e-03 8.6e-03 70 3.2e-01 2.8e-03 9.1e-03 69 3.1e-01 2.5e-03 8.9e-03 64 67 3.1e-01 2.5e-03 8.7e-03 63 2.9e-01 2.4e-03 8.2e-03 69 3.1e-01 2.5e-03 8.9e-03 128 67 3.2e-01 2.6e-03 8.9e-03 66 3.1e-01 2.5e-03 8.9e-03 68 3.2e-01 2.6e-03 9.0e-03 256 65 3.2e-01 2.6e-03 9.1e-03 65 3.1e-01 2.4e-03 8.7e-03 64 3.1e-01 2.5e-03 8.9e-03 512 63 3.1e-01 2.5e-03 8.9e-03 66 3.2e-01 2.5e-03 9.0e-03 64 3.2e-01 2.6e-03 9.1e-03 1024 62 3.2e-01 2.6e-03 8.9e-03 65 3.2e-01 2.6e-03 9.1e-03 60 3.0e-01 2.4e-03 8.8e-03 2048 59 3.8e-01 2.6e-03 1.0e-02 63 4.1e-01 2.9e-03 1.1e-02 62 4.0e-01 2.7e-03 1.1e-02 4096 57 4.4e-01 2.9e-03 1.2e-02 54 4.3e-01 2.9e-03 1.1e-02 56 4.4e-01 2.9e-03 1.2e-02 10624 38 4.1e-01 2.5e-03 1.3e-02 36 3.9e-01 2.5e-03 1.1e-02 37 4.0e-01 2.7e-03 1.0e-02 27554 28 4.9e-01 3.2e-03 1.8e-02 27 4.7e-01 2.9e-03 1.3e-02 26 4.5e-01 2.8e-03 1.2e-02 71468 16 6.0e-01 3.1e-03 2.1e-02 17 6.3e-01 3.6e-03 1.9e-02 18 6.6e-01 3.8e-03 2.0e-02 185364 9 6.5e-01 3.6e-03 2.2e-02 9 6.4e-01 3.8e-03 1.9e-02 9 6.3e-01 3.5e-03 2.0e-02 480774 4 6.7e-01 3.8e-03 2.1e-02 4 6.8e-01 3.5e-03 2.2e-02 4 6.7e-01 3.6e-03 2.1e-02 1246974 2 8.1e-01 4.3e-03 2.9e-02 2 8.2e-01 4.3e-03 3.0e-02 2 8.0e-01 4.4e-03 2.7e-02 3234251 1 3.6e-01 6.2e-03 3.5e-02 M 1 3.0e-01 6.4e-03 2.9e-02 M 1 2.7e-01 6.2e-03 3.1e-02 M 8388608 1 8.6e-01 1.6e-02 7.1e-02 R 1 7.2e-01 1.6e-02 7.1e-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 SECTION-LOOPLNGS-END SECTION-ELAPSED-BEGIN measurment loop: elapsed time = 87.059 sec sum of max elapsed time per entries above = 86.554 sec difference to elapsed time = 0.505 sec = 0.6% sum based on fastest repetition = 81.957 sec difference to elapsed time = 5.102 sec = 5.9% 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-4*2fix 1 8 1.00 1.00 0 ( 2 2 0 ) p01 ring-2*4fix 2 16 2.00 1.00 0 ( 1 0 0 ) p02 ring-1*8fix 2 16 2.00 1.00 0 ( 2 0 0 ) p03 ring-1*8fix 2 16 2.00 1.00 0 ( 0 0 0 ) p04 ring-1*8fix 2 16 2.00 1.00 0 ( 0 0 0 ) p05 ring-1*8fix 2 16 2.00 1.00 0 ( 0 0 0 ) p06 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 2 2 ) p07 random-cyc-1dim 2 16 2.00 1.00 0 ( 1 1 1 ) p08 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 0 0 ) p09 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 1 1 ) p10 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 2 0 ) p11 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 2 0 ) p12 random-cyc-1dim 2 16 2.00 1.00 0 ( 1 0 1 ) p13 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 1 1 ) p14 random-cyc-1dim 2 16 2.00 1.00 0 ( 1 0 0 ) p15 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 0 2 ) p16 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 2 0 ) p17 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 2 ) p18 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 2 ) p19 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 0 0 ) p20 random-cyc-1dim 2 16 2.00 1.00 0 ( 1 0 0 ) p21 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 1 1 ) p22 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 2 2 ) p23 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 2 0 ) p24 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 0 ) p25 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 0 ) p26 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 0 ) p27 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 0 0 ) p28 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 0 ) p29 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 0 ) p30 random-cyc-1dim 2 16 2.00 1.00 0 ( 2 2 2 ) p31 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 0 ) p32 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 0 ) p33 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 0 ) p34 random-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 2 ) p35 random-cyc-1dim 2 16 2.00 1.00 0 ( 1 0 1 ) p36 worst-cyc-1dim 2 16 2.00 1.00 0 ( 0 0 0 ) p37 best bi-section 2 8 1.00 0.50 0 ( 2 2 2 ) p38 worst bi-section 2 8 1.00 0.50 0 ( 0 0 0 ) p39 one PingPong Pair 2 2 1.00 0.50 6 ( 0 0 0 ) p40 acyclic-2dim-all 4 20 2.50 0.63 0 ( 0 2 0 ) p41 acyclic-3dim-all 6 24 3.00 0.50 0 ( 2 2 2 ) p42 cyclic-2dim-x 2 16 2.00 1.00 0 ( 0 2 2 ) p43 cyclic-2dim-y 1 8 1.00 1.00 0 ( 2 2 2 ) p44 cyclic-2dim-all 3 24 3.00 1.00 0 ( 2 2 2 ) p45 cyclic-3dim-x 1 8 1.00 1.00 0 ( 2 2 2 ) p46 cyclic-3dim-y 1 8 1.00 1.00 0 ( 2 2 2 ) p47 cyclic-3dim-z 1 8 1.00 1.00 0 ( 2 0 0 ) p48 cyclic-3dim-all 3 24 3.00 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-4*2fix : 77.420 56.543 70.791 -> 77.420 -> 619.357 MByte/s p01 ring-2*4fix : 71.732 64.548 69.416 -> 71.732 -> 573.858 MByte/s p02 ring-1*8fix : 70.206 61.215 69.192 -> 70.206 -> 561.650 MByte/s p03 ring-1*8fix : 69.970 61.197 68.982 -> 69.970 -> 559.764 MByte/s p04 ring-1*8fix : 72.108 61.543 69.446 -> 72.108 -> 576.861 MByte/s p05 ring-1*8fix : 69.991 61.109 69.058 -> 69.991 -> 559.930 MByte/s p06 random-cyc-1dim : 105.244 83.033 100.538 -> 105.244 -> 841.949 MByte/s p07 random-cyc-1dim : 77.901 63.959 76.789 -> 77.901 -> 623.207 MByte/s p08 random-cyc-1dim : 103.607 80.886 102.189 -> 103.607 -> 828.852 MByte/s p09 random-cyc-1dim : 81.692 66.871 81.137 -> 81.692 -> 653.533 MByte/s p10 random-cyc-1dim : 69.631 60.898 68.630 -> 69.631 -> 557.045 MByte/s p11 random-cyc-1dim : 109.366 83.995 101.677 -> 109.366 -> 874.931 MByte/s p12 random-cyc-1dim : 82.185 68.546 81.491 -> 82.185 -> 657.481 MByte/s p13 random-cyc-1dim : 83.610 70.123 82.805 -> 83.610 -> 668.884 MByte/s p14 random-cyc-1dim : 84.812 70.149 85.580 -> 85.580 -> 684.644 MByte/s p15 random-cyc-1dim : 103.617 82.134 103.501 -> 103.617 -> 828.937 MByte/s p16 random-cyc-1dim : 103.775 80.460 101.266 -> 103.775 -> 830.198 MByte/s p17 random-cyc-1dim : 82.127 69.336 82.924 -> 82.924 -> 663.392 MByte/s p18 random-cyc-1dim : 83.674 69.293 84.415 -> 84.415 -> 675.317 MByte/s p19 random-cyc-1dim : 135.129 95.779 130.664 -> 135.129 -> 1081.034 MByte/s p20 random-cyc-1dim : 70.016 61.132 70.064 -> 70.064 -> 560.508 MByte/s p21 random-cyc-1dim : 81.804 67.382 81.990 -> 81.990 -> 655.917 MByte/s p22 random-cyc-1dim : 131.824 94.583 129.382 -> 131.824 -> 1054.591 MByte/s p23 random-cyc-1dim : 110.706 84.943 104.854 -> 110.706 -> 885.650 MByte/s p24 random-cyc-1dim : 83.212 70.318 82.799 -> 83.212 -> 665.694 MByte/s p25 random-cyc-1dim : 135.547 95.968 128.799 -> 135.547 -> 1084.376 MByte/s p26 random-cyc-1dim : 112.595 87.084 106.216 -> 112.595 -> 900.764 MByte/s p27 random-cyc-1dim : 112.582 87.071 105.755 -> 112.582 -> 900.655 MByte/s p28 random-cyc-1dim : 113.077 84.812 105.914 -> 113.077 -> 904.612 MByte/s p29 random-cyc-1dim : 113.534 86.962 106.148 -> 113.534 -> 908.274 MByte/s p30 random-cyc-1dim : 103.947 81.431 101.332 -> 103.947 -> 831.573 MByte/s p31 random-cyc-1dim : 84.047 69.295 83.798 -> 84.047 -> 672.376 MByte/s p32 random-cyc-1dim : 82.899 69.129 83.095 -> 83.095 -> 664.762 MByte/s p33 random-cyc-1dim : 112.048 84.251 102.613 -> 112.048 -> 896.386 MByte/s p34 random-cyc-1dim : 110.166 84.629 107.160 -> 110.166 -> 881.326 MByte/s p35 random-cyc-1dim : 82.407 68.794 83.018 -> 83.018 -> 664.145 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 110.988 80.495 102.497 -> 110.988 -> 887.904 MByte/s p37 best bi-section : 59.624 56.184 72.154 -> 72.154 -> 577.232 MByte/s p38 worst bi-section : 67.645 59.874 72.690 -> 72.690 -> 581.523 MByte/s p39 one PingPong Pair : 25.350 9.143 9.143 -> 25.350 -> 202.798 MByte/s p40 acyclic-2dim-all : 87.558 84.367 104.968 -> 104.968 -> 839.742 MByte/s p41 acyclic-3dim-all : 94.431 101.814 120.623 -> 120.623 -> 964.980 MByte/s p42 cyclic-2dim-x : 167.677 103.274 158.846 -> 167.677 -> 1341.419 MByte/s p43 cyclic-2dim-y : 76.519 55.796 69.074 -> 76.519 -> 612.149 MByte/s p44 cyclic-2dim-all : 121.714 88.692 121.527 -> 121.714 -> 973.711 MByte/s p45 cyclic-3dim-x : 177.519 78.231 156.223 -> 177.519 -> 1420.149 MByte/s p46 cyclic-3dim-y : 179.633 100.466 165.146 -> 179.633 -> 1437.065 MByte/s p47 cyclic-3dim-z : 76.662 55.287 69.995 -> 76.662 -> 613.292 MByte/s p48 cyclic-3dim-all : 118.457 101.767 121.812 -> 121.812 -> 974.492 MByte/s log_avg of all rings : 71.859 60.980 69.478 || 71.859 -> 574.871 MByte/s log_avg of all random : 96.539 76.774 94.212 || 96.668 -> 773.345 MByte/s log_avg(ring,random) : 83.290 68.423 80.905 ||( 83.345 -> 666.763)MByte/s * size -> accumulated on all pr.: 666.317 547.382 647.243 ||(666.763)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-4*2fix : 73.827 76.491 74.939 -> 76.491 -> 611.927 MByte/s p01 ring-2*4fix : 67.515 71.487 72.154 -> 72.154 -> 577.228 MByte/s p02 ring-1*8fix : 66.777 71.130 70.686 -> 71.130 -> 569.039 MByte/s p03 ring-1*8fix : 70.456 70.169 70.651 -> 70.651 -> 565.210 MByte/s p04 ring-1*8fix : 73.142 70.078 70.946 -> 73.142 -> 585.136 MByte/s p05 ring-1*8fix : 70.795 71.390 70.596 -> 71.390 -> 571.124 MByte/s p06 random-cyc-1dim : 96.726 103.319 102.822 -> 103.319 -> 826.548 MByte/s p07 random-cyc-1dim : 72.419 78.403 80.456 -> 80.456 -> 643.645 MByte/s p08 random-cyc-1dim : 95.130 103.769 104.493 -> 104.493 -> 835.941 MByte/s p09 random-cyc-1dim : 81.741 79.227 82.501 -> 82.501 -> 660.007 MByte/s p10 random-cyc-1dim : 69.146 69.746 71.038 -> 71.038 -> 568.303 MByte/s p11 random-cyc-1dim : 101.711 107.319 106.128 -> 107.319 -> 858.553 MByte/s p12 random-cyc-1dim : 76.001 82.342 80.826 -> 82.342 -> 658.734 MByte/s p13 random-cyc-1dim : 85.162 79.626 81.598 -> 85.162 -> 681.300 MByte/s p14 random-cyc-1dim : 76.714 85.661 84.378 -> 85.661 -> 685.290 MByte/s p15 random-cyc-1dim : 101.116 103.459 103.373 -> 103.459 -> 827.672 MByte/s p16 random-cyc-1dim : 103.341 102.335 104.487 -> 104.487 -> 835.893 MByte/s p17 random-cyc-1dim : 83.078 84.538 83.470 -> 84.538 -> 676.307 MByte/s p18 random-cyc-1dim : 85.001 86.228 85.412 -> 86.228 -> 689.824 MByte/s p19 random-cyc-1dim : 131.736 131.924 128.847 -> 131.924 -> 1055.393 MByte/s p20 random-cyc-1dim : 69.886 70.339 69.938 -> 70.339 -> 562.711 MByte/s p21 random-cyc-1dim : 84.537 78.674 82.501 -> 84.537 -> 676.297 MByte/s p22 random-cyc-1dim : 130.515 128.146 132.039 -> 132.039 -> 1056.309 MByte/s p23 random-cyc-1dim : 110.528 106.864 109.711 -> 110.528 -> 884.226 MByte/s p24 random-cyc-1dim : 84.161 84.964 85.954 -> 85.954 -> 687.629 MByte/s p25 random-cyc-1dim : 128.856 126.393 134.643 -> 134.643 -> 1077.146 MByte/s p26 random-cyc-1dim : 110.263 109.062 108.086 -> 110.263 -> 882.103 MByte/s p27 random-cyc-1dim : 105.394 110.904 110.035 -> 110.904 -> 887.235 MByte/s p28 random-cyc-1dim : 112.673 109.459 106.086 -> 112.673 -> 901.384 MByte/s p29 random-cyc-1dim : 110.525 108.092 111.267 -> 111.267 -> 890.134 MByte/s p30 random-cyc-1dim : 103.681 100.501 104.600 -> 104.600 -> 836.800 MByte/s p31 random-cyc-1dim : 86.342 84.703 84.588 -> 86.342 -> 690.739 MByte/s p32 random-cyc-1dim : 84.974 83.575 85.437 -> 85.437 -> 683.496 MByte/s p33 random-cyc-1dim : 109.069 108.708 111.279 -> 111.279 -> 890.231 MByte/s p34 random-cyc-1dim : 109.316 107.452 106.016 -> 109.316 -> 874.528 MByte/s p35 random-cyc-1dim : 78.884 84.704 79.229 -> 84.704 -> 677.634 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 108.434 109.432 105.117 -> 109.432 -> 875.453 MByte/s p37 best bi-section : 70.336 68.988 69.420 -> 70.336 -> 562.691 MByte/s p38 worst bi-section : 71.419 71.379 70.831 -> 71.419 -> 571.353 MByte/s p39 one PingPong Pair : 25.008 24.980 24.674 -> 25.008 -> 200.061 MByte/s p40 acyclic-2dim-all : 91.814 103.148 91.984 -> 103.148 -> 825.183 MByte/s p41 acyclic-3dim-all : 115.782 118.509 120.178 -> 120.178 -> 961.425 MByte/s p42 cyclic-2dim-x : 160.787 161.339 167.764 -> 167.764 -> 1342.110 MByte/s p43 cyclic-2dim-y : 75.050 75.169 76.121 -> 76.121 -> 608.972 MByte/s p44 cyclic-2dim-all : 123.918 122.707 119.953 -> 123.918 -> 991.343 MByte/s p45 cyclic-3dim-x : 174.511 175.457 177.930 -> 177.930 -> 1423.439 MByte/s p46 cyclic-3dim-y : 179.417 166.848 180.483 -> 180.483 -> 1443.863 MByte/s p47 cyclic-3dim-z : 75.895 75.483 75.174 -> 75.895 -> 607.157 MByte/s p48 cyclic-3dim-all : 122.791 120.158 122.548 -> 122.791 -> 982.324 MByte/s log_avg of all rings : 70.370 71.759 71.645 || 72.467 -> 579.737 MByte/s log_avg of all random : 94.390 95.270 95.900 || 97.131 -> 777.051 MByte/s log_avg(ring,random) : 81.500 82.683 82.890 ||( 83.898 -> 671.182)MByte/s * size -> accumulated on all pr.: 652.000 661.463 663.124 ||(671.182)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-4*2fix p00 method 0 =Sndrcv :( 28.204) 0.035 0.558 8.125 55.232 212.613 232.856 -> 77.420 -> 619.357 MByte/s p00 method 1 =Alltoal :(170.764) 0.006 0.093 1.455 16.681 142.586 232.856 -> 56.543 -> 452.342 MByte/s p00 method 2 =non-blk :( 52.700) 0.019 0.287 4.509 44.514 184.218 232.856 -> 70.791 -> 566.326 MByte/s p01 ring-2*4fix p01 method 0 =Sndrcv :( 28.185) 0.035 0.551 7.944 55.332 198.729 229.360 -> 71.732 -> 573.858 MByte/s p01 method 1 =Alltoal :( 86.001) 0.012 0.183 2.813 27.966 177.296 229.360 -> 64.548 -> 516.387 MByte/s p01 method 2 =non-blk :( 48.832) 0.020 0.314 4.636 46.123 182.645 229.360 -> 69.416 -> 555.328 MByte/s p02 ring-1*8fix p02 method 0 =Sndrcv :( 28.460) 0.035 0.538 7.897 57.861 166.069 228.017 -> 70.206 -> 561.650 MByte/s p02 method 1 =Alltoal :( 85.685) 0.012 0.184 2.830 24.566 161.867 228.017 -> 61.215 -> 489.721 MByte/s p02 method 2 =non-blk :( 49.423) 0.020 0.309 4.616 46.154 182.894 228.017 -> 69.192 -> 553.538 MByte/s p03 ring-1*8fix p03 method 0 =Sndrcv :( 28.352) 0.035 0.540 7.779 58.995 169.760 229.853 -> 69.970 -> 559.764 MByte/s p03 method 1 =Alltoal :( 86.056) 0.012 0.184 2.818 24.721 156.912 229.853 -> 61.197 -> 489.574 MByte/s p03 method 2 =non-blk :( 49.747) 0.020 0.310 4.674 45.873 172.744 229.853 -> 68.982 -> 551.859 MByte/s p04 ring-1*8fix p04 method 0 =Sndrcv :( 28.257) 0.035 0.540 7.755 59.165 191.079 231.944 -> 72.108 -> 576.861 MByte/s p04 method 1 =Alltoal :( 86.418) 0.012 0.181 2.837 24.704 161.797 231.944 -> 61.543 -> 492.344 MByte/s p04 method 2 =non-blk :( 49.725) 0.020 0.310 4.653 45.743 182.585 231.944 -> 69.446 -> 555.572 MByte/s p05 ring-1*8fix p05 method 0 =Sndrcv :( 28.270) 0.035 0.542 7.883 59.033 173.231 228.106 -> 69.991 -> 559.930 MByte/s p05 method 1 =Alltoal :( 86.322) 0.012 0.181 2.817 24.686 159.796 228.106 -> 61.109 -> 488.872 MByte/s p05 method 2 =non-blk :( 49.795) 0.020 0.308 4.640 45.216 179.945 228.106 -> 69.058 -> 552.462 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 27.508) 0.036 0.568 8.228 60.279 282.657 358.557 -> 105.244 -> 841.949 MByte/s p06 method 1 =Alltoal :( 95.758) 0.010 0.161 2.591 29.940 197.474 358.557 -> 83.033 -> 664.266 MByte/s p06 method 2 =non-blk :( 47.447) 0.021 0.327 4.764 46.472 256.127 358.557 -> 100.538 -> 804.301 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 27.822) 0.036 0.554 7.972 56.889 198.942 227.066 -> 77.901 -> 623.207 MByte/s p07 method 1 =Alltoal :( 89.523) 0.011 0.172 2.724 26.873 168.181 227.066 -> 63.959 -> 511.674 MByte/s p07 method 2 =non-blk :( 48.427) 0.021 0.320 4.703 46.013 213.812 227.066 -> 76.789 -> 614.314 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 27.633) 0.036 0.565 8.198 59.656 231.163 369.079 -> 103.607 -> 828.852 MByte/s p08 method 1 =Alltoal :( 95.990) 0.010 0.161 2.571 30.186 167.094 369.079 -> 80.886 -> 647.090 MByte/s p08 method 2 =non-blk :( 47.730) 0.021 0.330 4.711 46.991 257.529 369.079 -> 102.189 -> 817.515 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 27.857) 0.036 0.557 7.997 58.846 209.186 267.759 -> 81.692 -> 653.533 MByte/s p09 method 1 =Alltoal :( 89.402) 0.011 0.172 2.687 27.112 170.113 267.759 -> 66.871 -> 534.970 MByte/s p09 method 2 =non-blk :( 48.456) 0.021 0.317 4.735 45.743 214.073 267.759 -> 81.137 -> 649.094 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 28.356) 0.035 0.538 7.905 54.954 174.939 230.494 -> 69.631 -> 557.045 MByte/s p10 method 1 =Alltoal :( 85.372) 0.012 0.184 2.830 25.423 157.668 230.494 -> 60.898 -> 487.187 MByte/s p10 method 2 =non-blk :( 49.214) 0.020 0.310 4.668 44.989 179.105 230.494 -> 68.630 -> 549.037 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 27.588) 0.036 0.568 8.210 60.134 277.942 362.421 -> 109.366 -> 874.931 MByte/s p11 method 1 =Alltoal :( 94.648) 0.011 0.168 2.589 29.451 200.608 362.421 -> 83.995 -> 671.961 MByte/s p11 method 2 =non-blk :( 47.170) 0.021 0.326 4.839 47.362 258.908 362.421 -> 101.677 -> 813.416 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 27.777) 0.036 0.550 8.054 59.418 197.425 280.335 -> 82.185 -> 657.481 MByte/s p12 method 1 =Alltoal :( 89.308) 0.011 0.178 2.723 26.690 170.608 280.335 -> 68.546 -> 548.367 MByte/s p12 method 2 =non-blk :( 49.310) 0.020 0.316 4.738 46.073 218.806 280.335 -> 81.491 -> 651.928 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 27.680) 0.036 0.553 8.082 60.010 197.853 295.088 -> 83.610 -> 668.884 MByte/s p13 method 1 =Alltoal :( 89.830) 0.011 0.178 2.728 26.386 160.004 295.088 -> 70.123 -> 560.985 MByte/s p13 method 2 =non-blk :( 48.714) 0.021 0.315 4.776 46.218 212.723 295.088 -> 82.805 -> 662.440 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 27.672) 0.036 0.560 8.192 59.721 194.943 294.084 -> 84.812 -> 678.495 MByte/s p14 method 1 =Alltoal :( 89.968) 0.011 0.177 2.702 26.610 168.233 294.084 -> 70.149 -> 561.192 MByte/s p14 method 2 =non-blk :( 47.925) 0.021 0.316 4.697 46.100 227.178 294.084 -> 85.580 -> 684.644 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 27.502) 0.036 0.564 8.222 60.502 254.445 366.410 -> 103.617 -> 828.937 MByte/s p15 method 1 =Alltoal :( 95.600) 0.010 0.167 2.570 29.990 187.584 366.410 -> 82.134 -> 657.072 MByte/s p15 method 2 =non-blk :( 47.558) 0.021 0.323 4.793 45.990 266.456 366.410 -> 103.501 -> 828.011 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 27.553) 0.036 0.566 8.162 59.765 246.058 354.690 -> 103.775 -> 830.198 MByte/s p16 method 1 =Alltoal :( 95.524) 0.010 0.167 2.563 30.186 184.075 354.690 -> 80.460 -> 643.677 MByte/s p16 method 2 =non-blk :( 47.500) 0.021 0.327 4.874 45.914 247.463 354.690 -> 101.266 -> 810.126 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 27.898) 0.036 0.554 8.067 58.829 201.492 295.624 -> 82.127 -> 657.017 MByte/s p17 method 1 =Alltoal :( 89.472) 0.011 0.177 2.711 27.244 172.484 295.624 -> 69.336 -> 554.691 MByte/s p17 method 2 =non-blk :( 48.519) 0.021 0.314 4.820 46.005 213.021 295.624 -> 82.924 -> 663.392 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 27.793) 0.036 0.560 8.124 59.761 188.729 290.409 -> 83.674 -> 669.389 MByte/s p18 method 1 =Alltoal :( 89.830) 0.011 0.178 2.706 27.441 158.251 290.409 -> 69.293 -> 554.343 MByte/s p18 method 2 =non-blk :( 48.257) 0.021 0.319 4.800 46.685 215.916 290.409 -> 84.415 -> 675.317 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 24.838) 0.040 0.600 8.808 73.549 311.558 485.648 -> 135.129 -> 1081.034 MByte/s p19 method 1 =Alltoal :(101.335) 0.010 0.158 2.462 33.701 203.438 485.648 -> 95.779 -> 766.231 MByte/s p19 method 2 =non-blk :( 44.044) 0.023 0.349 5.336 55.565 327.307 485.648 -> 130.664 -> 1045.312 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 28.213) 0.035 0.540 7.755 58.623 170.391 231.212 -> 70.016 -> 560.129 MByte/s p20 method 1 =Alltoal :( 85.630) 0.012 0.184 2.810 25.435 162.173 231.212 -> 61.132 -> 489.057 MByte/s p20 method 2 =non-blk :( 49.603) 0.020 0.307 4.512 46.200 179.374 231.212 -> 70.064 -> 560.508 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 27.577) 0.036 0.564 8.050 60.001 201.721 287.443 -> 81.804 -> 654.431 MByte/s p21 method 1 =Alltoal :( 89.452) 0.011 0.177 2.713 27.335 176.648 287.443 -> 67.382 -> 539.054 MByte/s p21 method 2 =non-blk :( 48.045) 0.021 0.324 4.772 46.430 223.585 287.443 -> 81.990 -> 655.917 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 24.615) 0.041 0.608 8.846 72.798 325.200 486.226 -> 131.824 -> 1054.591 MByte/s p22 method 1 =Alltoal :(101.245) 0.010 0.156 2.462 33.135 198.779 486.226 -> 94.583 -> 756.664 MByte/s p22 method 2 =non-blk :( 44.325) 0.023 0.356 5.259 55.057 336.310 486.226 -> 129.382 -> 1035.056 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 27.835) 0.036 0.568 8.196 61.235 289.158 369.575 -> 110.706 -> 885.650 MByte/s p23 method 1 =Alltoal :( 95.535) 0.010 0.164 2.566 29.573 186.485 369.575 -> 84.943 -> 679.547 MByte/s p23 method 2 =non-blk :( 47.615) 0.021 0.328 4.919 47.823 257.032 369.575 -> 104.854 -> 838.828 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 28.007) 0.036 0.552 8.079 59.405 195.111 288.372 -> 83.212 -> 665.694 MByte/s p24 method 1 =Alltoal :( 89.546) 0.011 0.174 2.728 27.335 172.322 288.372 -> 70.318 -> 562.546 MByte/s p24 method 2 =non-blk :( 49.145) 0.020 0.318 4.764 46.448 215.471 288.372 -> 82.799 -> 662.391 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 24.907) 0.040 0.602 8.994 72.831 347.694 489.090 -> 135.547 -> 1084.376 MByte/s p25 method 1 =Alltoal :(101.263) 0.010 0.158 2.422 33.332 203.212 489.090 -> 95.968 -> 767.742 MByte/s p25 method 2 =non-blk :( 44.588) 0.022 0.348 5.304 54.748 317.891 489.090 -> 128.799 -> 1030.394 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 27.433) 0.036 0.570 8.358 61.483 274.546 404.612 -> 112.595 -> 900.764 MByte/s p26 method 1 =Alltoal :( 94.703) 0.011 0.166 2.554 29.649 200.070 404.612 -> 87.084 -> 696.672 MByte/s p26 method 2 =non-blk :( 46.738) 0.021 0.329 5.032 48.219 254.446 404.612 -> 106.216 -> 849.731 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 27.408) 0.036 0.572 8.342 60.800 278.029 400.116 -> 112.582 -> 900.655 MByte/s p27 method 1 =Alltoal :( 94.950) 0.011 0.167 2.557 29.346 212.086 400.116 -> 87.071 -> 696.565 MByte/s p27 method 2 =non-blk :( 46.776) 0.021 0.329 4.974 47.792 271.022 400.116 -> 105.755 -> 846.042 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 27.439) 0.036 0.570 8.312 61.189 285.578 397.027 -> 113.077 -> 904.612 MByte/s p28 method 1 =Alltoal :( 95.475) 0.010 0.167 2.583 29.387 179.877 397.027 -> 84.812 -> 678.494 MByte/s p28 method 2 =non-blk :( 46.738) 0.021 0.327 4.936 47.976 257.966 397.027 -> 105.914 -> 847.311 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 27.425) 0.036 0.570 8.283 61.098 275.784 394.470 -> 113.534 -> 908.274 MByte/s p29 method 1 =Alltoal :( 95.648) 0.010 0.166 2.576 29.394 208.470 394.470 -> 86.962 -> 695.697 MByte/s p29 method 2 =non-blk :( 46.626) 0.021 0.328 4.972 47.619 266.646 394.470 -> 106.148 -> 849.188 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 27.393) 0.037 0.570 8.115 60.637 259.176 364.516 -> 103.947 -> 831.573 MByte/s p30 method 1 =Alltoal :( 96.087) 0.010 0.166 2.577 29.931 192.452 364.516 -> 81.431 -> 651.445 MByte/s p30 method 2 =non-blk :( 47.418) 0.021 0.326 4.803 47.609 264.492 364.516 -> 101.332 -> 810.654 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 27.740) 0.036 0.563 8.065 59.548 220.519 286.947 -> 84.047 -> 672.376 MByte/s p31 method 1 =Alltoal :( 89.720) 0.011 0.177 2.726 27.030 170.477 286.947 -> 69.295 -> 554.357 MByte/s p31 method 2 =non-blk :( 47.907) 0.021 0.322 4.777 46.498 216.013 286.947 -> 83.798 -> 670.386 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 27.750) 0.036 0.562 8.094 59.721 194.141 284.644 -> 82.899 -> 663.191 MByte/s p32 method 1 =Alltoal :( 89.758) 0.011 0.177 2.700 26.544 168.130 284.644 -> 69.129 -> 553.035 MByte/s p32 method 2 =non-blk :( 47.980) 0.021 0.324 4.649 45.946 211.093 284.644 -> 83.095 -> 664.762 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 27.593) 0.036 0.567 8.179 61.080 281.546 402.776 -> 112.048 -> 896.386 MByte/s p33 method 1 =Alltoal :( 95.140) 0.011 0.166 2.591 28.947 189.339 402.776 -> 84.251 -> 674.006 MByte/s p33 method 2 =non-blk :( 47.155) 0.021 0.324 4.815 47.782 265.228 402.776 -> 102.613 -> 820.904 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 27.496) 0.036 0.572 8.266 61.384 259.399 381.101 -> 110.166 -> 881.326 MByte/s p34 method 1 =Alltoal :( 95.693) 0.010 0.164 2.586 29.362 197.721 381.101 -> 84.629 -> 677.028 MByte/s p34 method 2 =non-blk :( 47.059) 0.021 0.333 4.943 48.751 278.254 381.101 -> 107.160 -> 857.282 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 27.608) 0.036 0.554 8.099 58.157 199.405 288.249 -> 82.407 -> 659.252 MByte/s p35 method 1 =Alltoal :( 90.050) 0.011 0.172 2.706 27.023 173.211 288.249 -> 68.794 -> 550.353 MByte/s p35 method 2 =non-blk :( 48.713) 0.021 0.320 4.788 46.000 221.360 288.249 -> 83.018 -> 664.145 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 24.902) 0.040 0.596 8.935 72.234 286.553 363.781 -> 110.988 -> 887.904 MByte/s p36 method 1 =Alltoal :( 92.787) 0.011 0.167 2.608 28.651 179.645 363.781 -> 80.495 -> 643.959 MByte/s p36 method 2 =non-blk :( 45.961) 0.022 0.337 5.169 53.444 263.300 363.781 -> 102.497 -> 819.974 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 24.503) 0.020 0.303 4.405 35.839 169.166 232.307 -> 59.624 -> 476.995 MByte/s p37 method 1 =Alltoal :( 85.056) 0.006 0.093 1.410 16.708 144.480 232.307 -> 56.184 -> 449.473 MByte/s p37 method 2 =non-blk :( 25.937) 0.019 0.287 4.549 44.269 207.421 232.307 -> 72.154 -> 577.232 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 24.580) 0.020 0.302 4.421 36.944 181.314 234.732 -> 67.645 -> 541.158 MByte/s p38 method 1 =Alltoal :( 85.932) 0.006 0.093 1.438 18.644 160.906 234.732 -> 59.874 -> 478.992 MByte/s p38 method 2 =non-blk :( 26.417) 0.019 0.293 4.538 44.051 208.559 234.732 -> 72.690 -> 581.523 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 23.430) 0.005 0.080 1.145 9.887 59.239 96.987 -> 25.350 -> 202.798 MByte/s p39 method 1 =Alltoal :(-------) 0.000 0.000 0.000 0.000 0.000 96.987 -> 9.143 -> 73.143 MByte/s p39 method 2 =non-blk :(-------) 0.000 0.000 0.000 0.000 0.000 96.987 -> 9.143 -> 73.143 MByte/s p40 acyclic-2dim-all p40 method 0 =Sndrcv :( 19.883) 0.031 0.470 6.783 57.156 204.566 339.098 -> 87.558 -> 700.467 MByte/s p40 method 1 =Alltoal :( 43.562) 0.014 0.228 3.441 36.203 207.500 339.098 -> 84.367 -> 674.934 MByte/s p40 method 2 =non-blk :( 29.527) 0.021 0.329 4.952 53.353 284.613 339.098 -> 104.968 -> 839.742 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 16.059) 0.031 0.471 6.769 59.446 209.508 413.639 -> 94.431 -> 755.450 MByte/s p41 method 1 =Alltoal :( 29.581) 0.017 0.269 4.033 41.719 251.551 413.639 -> 101.814 -> 814.509 MByte/s p41 method 2 =non-blk :( 20.011) 0.025 0.390 5.913 61.879 322.207 413.639 -> 120.623 -> 964.980 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 15.460) 0.065 1.020 14.523 129.498 385.743 517.048 -> 167.677 -> 1341.419 MByte/s p42 method 1 =Alltoal :( 86.263) 0.012 0.185 2.848 35.463 222.750 517.048 -> 103.274 -> 826.195 MByte/s p42 method 2 =non-blk :( 34.529) 0.029 0.459 6.779 82.523 400.593 517.048 -> 158.846 -> 1270.765 MByte/s p43 cyclic-2dim-y p43 method 0 =Sndrcv :( 28.137) 0.036 0.561 8.149 56.754 215.476 229.643 -> 76.519 -> 612.149 MByte/s p43 method 1 =Alltoal :(170.623) 0.006 0.093 1.451 16.364 140.923 229.643 -> 55.796 -> 446.366 MByte/s p43 method 2 =non-blk :( 52.887) 0.019 0.290 4.403 44.591 192.288 229.643 -> 69.074 -> 552.591 MByte/s p44 cyclic-2dim-all p44 method 0 =Sndrcv :( 19.548) 0.051 0.782 11.279 90.600 307.900 416.605 -> 121.714 -> 973.711 MByte/s p44 method 1 =Alltoal :( 58.023) 0.017 0.275 4.085 42.027 191.845 416.605 -> 88.692 -> 709.535 MByte/s p44 method 2 =non-blk :( 41.021) 0.024 0.378 5.721 61.937 320.862 416.605 -> 121.527 -> 972.214 MByte/s p45 cyclic-3dim-x p45 method 0 =Sndrcv :( 15.521) 0.064 1.015 14.576 126.299 416.378 519.580 -> 177.519 -> 1420.149 MByte/s p45 method 1 =Alltoal :(171.753) 0.006 0.094 1.460 18.349 106.715 519.580 -> 78.231 -> 625.846 MByte/s p45 method 2 =non-blk :( 35.320) 0.028 0.449 6.825 79.874 363.381 519.580 -> 156.223 -> 1249.783 MByte/s p46 cyclic-3dim-y p46 method 0 =Sndrcv :( 15.687) 0.064 1.006 14.396 125.957 432.590 519.001 -> 179.633 -> 1437.065 MByte/s p46 method 1 =Alltoal :(171.680) 0.006 0.094 1.468 19.795 213.719 519.001 -> 100.466 -> 803.731 MByte/s p46 method 2 =non-blk :( 35.874) 0.028 0.442 6.577 79.414 438.562 519.001 -> 165.146 -> 1321.170 MByte/s p47 cyclic-3dim-z p47 method 0 =Sndrcv :( 28.187) 0.035 0.561 8.182 58.960 213.104 221.956 -> 76.662 -> 613.292 MByte/s p47 method 1 =Alltoal :(171.323) 0.006 0.094 1.460 16.776 143.693 221.956 -> 55.287 -> 442.297 MByte/s p47 method 2 =non-blk :( 52.697) 0.019 0.287 4.447 44.933 175.111 221.956 -> 69.995 -> 559.963 MByte/s p48 cyclic-3dim-all p48 method 0 =Sndrcv :( 19.674) 0.051 0.784 11.336 87.935 288.254 401.113 -> 118.457 -> 947.654 MByte/s p48 method 1 =Alltoal :( 58.028) 0.017 0.275 4.188 43.116 253.381 401.113 -> 101.767 -> 814.134 MByte/s p48 method 2 =non-blk :( 40.952) 0.024 0.378 5.718 61.776 328.682 401.113 -> 121.812 -> 974.492 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.035 0.545 7.896 57.578 184.488 230.015 || 71.859 -> 574.871 MByte/s - ring, method 1 = Alltoal: 0.010 0.163 2.528 23.600 159.718 230.015 || 60.980 -> 487.842 MByte/s - ring, method 2 = non-blk: 0.020 0.306 4.621 45.600 180.797 230.015 || 69.478 -> 555.826 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.037 0.565 8.204 60.944 236.258 333.622 || 96.539 -> 772.309 MByte/s - random, method 1 = Alltoal: 0.011 0.170 2.632 28.588 181.210 333.622 || 76.774 -> 614.188 MByte/s - random, method 2 = non-blk: 0.021 0.325 4.846 47.498 241.980 333.622 || 94.212 -> 753.695 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.036 0.555 8.049 59.237 208.774 277.016 || 83.290 -> 666.317 MByte/s - average, method 1 = Alltoal: 0.011 0.166 2.579 25.975 170.125 277.016 || 68.423 -> 547.382 MByte/s - average, method 2 = non-blk: 0.021 0.315 4.732 46.539 209.163 277.016 || 80.905 -> 647.243 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 0.287 4.439 64.391 473.896 1670.195 2216.132 || 666.317 MByte/s - accumulated, mthd 1 = Alltoal: 0.084 1.331 20.633 207.797 1360.999 2216.132 || 547.382 MByte/s - accumulated, mthd 2 = non-blk: 0.164 2.524 37.855 372.315 1673.306 2216.132 || 647.243 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 0.287 0.036 0.035 0.037 0.036 0.011 0.021 2 0.573 0.072 0.071 0.073 0.072 0.021 0.041 4 1.133 0.142 0.139 0.144 0.142 0.042 0.081 8 2.293 0.287 0.282 0.292 0.287 0.085 0.165 16 4.439 0.555 0.545 0.565 0.555 0.166 0.315 32 8.769 1.096 1.078 1.115 1.096 0.334 0.624 64 17.218 2.152 2.114 2.192 2.152 0.664 1.238 128 33.186 4.148 4.084 4.214 4.148 1.310 2.428 256 64.391 8.049 7.896 8.204 8.049 2.579 4.732 512 127.015 15.877 15.633 16.125 15.877 5.101 9.286 1024 247.290 30.911 30.514 31.314 30.911 10.111 18.305 2048 297.864 37.233 36.662 37.813 37.233 15.131 27.963 4096 473.896 59.237 57.578 60.944 59.237 25.975 46.539 10624 683.065 85.383 78.754 92.571 79.814 48.657 83.707 27554 1080.789 135.099 123.206 148.139 119.719 81.909 131.888 71468 1263.722 157.965 135.443 184.232 152.477 125.554 155.628 185364 1731.762 216.470 189.220 247.645 208.774 170.125 209.163 480774 1881.283 235.160 201.477 274.475 231.949 203.049 223.615 1246974 2015.900 251.987 205.154 309.512 251.433 200.618 239.372 3234251 2105.308 263.163 220.057 314.715 263.163 263.163 263.163 8388608 2216.132 277.016 230.015 333.622 277.016 277.016 277.016 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-4*2fix :( 28.204) 0.035 0.558 8.125 55.232 212.613 232.856 -> 77.748 -> 621.983 MByte/s p01 ring-2*4fix :( 28.185) 0.035 0.551 7.944 55.332 198.729 229.360 -> 73.053 -> 584.423 MByte/s p02 ring-1*8fix :( 28.460) 0.035 0.538 7.897 57.861 182.894 228.017 -> 72.480 -> 579.840 MByte/s p03 ring-1*8fix :( 28.352) 0.035 0.540 7.779 58.995 172.744 229.853 -> 71.518 -> 572.145 MByte/s p04 ring-1*8fix :( 28.257) 0.035 0.540 7.755 59.165 191.079 231.944 -> 73.343 -> 586.741 MByte/s p05 ring-1*8fix :( 28.270) 0.035 0.542 7.883 59.033 179.945 228.106 -> 72.189 -> 577.513 MByte/s p06 random-cyc-1dim :( 27.508) 0.036 0.568 8.228 60.279 282.657 358.557 -> 106.404 -> 851.234 MByte/s p07 random-cyc-1dim :( 27.822) 0.036 0.554 7.972 56.889 213.812 227.066 -> 80.661 -> 645.289 MByte/s p08 random-cyc-1dim :( 27.633) 0.036 0.565 8.198 59.656 257.529 369.079 -> 105.891 -> 847.128 MByte/s p09 random-cyc-1dim :( 27.857) 0.036 0.557 7.997 58.846 214.073 267.759 -> 84.082 -> 672.655 MByte/s p10 random-cyc-1dim :( 28.356) 0.035 0.538 7.905 54.954 179.105 230.494 -> 71.210 -> 569.682 MByte/s p11 random-cyc-1dim :( 27.588) 0.036 0.568 8.210 60.134 277.942 362.421 -> 109.366 -> 874.931 MByte/s p12 random-cyc-1dim :( 27.777) 0.036 0.550 8.054 59.418 218.806 280.335 -> 85.454 -> 683.632 MByte/s p13 random-cyc-1dim :( 27.680) 0.036 0.553 8.082 60.010 212.723 295.088 -> 86.478 -> 691.822 MByte/s p14 random-cyc-1dim :( 27.672) 0.036 0.560 8.192 59.721 227.178 294.084 -> 88.670 -> 709.363 MByte/s p15 random-cyc-1dim :( 27.502) 0.036 0.564 8.222 60.502 266.456 366.410 -> 107.011 -> 856.088 MByte/s p16 random-cyc-1dim :( 27.553) 0.036 0.566 8.162 59.765 247.463 354.690 -> 105.101 -> 840.808 MByte/s p17 random-cyc-1dim :( 27.898) 0.036 0.554 8.067 58.829 213.021 295.624 -> 85.246 -> 681.967 MByte/s p18 random-cyc-1dim :( 27.793) 0.036 0.560 8.124 59.761 215.916 290.409 -> 87.678 -> 701.425 MByte/s p19 random-cyc-1dim :( 24.838) 0.040 0.600 8.808 73.549 327.307 485.648 -> 135.879 -> 1087.034 MByte/s p20 random-cyc-1dim :( 28.213) 0.035 0.540 7.755 58.623 179.374 231.212 -> 72.424 -> 579.390 MByte/s p21 random-cyc-1dim :( 27.577) 0.036 0.564 8.050 60.001 223.585 287.443 -> 85.087 -> 680.698 MByte/s p22 random-cyc-1dim :( 24.615) 0.041 0.608 8.846 72.798 336.310 486.226 -> 133.597 -> 1068.780 MByte/s p23 random-cyc-1dim :( 27.835) 0.036 0.568 8.196 61.235 289.158 369.575 -> 111.050 -> 888.400 MByte/s p24 random-cyc-1dim :( 28.007) 0.036 0.552 8.079 59.405 215.471 288.372 -> 86.867 -> 694.940 MByte/s p25 random-cyc-1dim :( 24.907) 0.040 0.602 8.994 72.831 347.694 489.090 -> 135.598 -> 1084.780 MByte/s p26 random-cyc-1dim :( 27.433) 0.036 0.570 8.358 61.483 274.546 404.612 -> 112.868 -> 902.943 MByte/s p27 random-cyc-1dim :( 27.408) 0.036 0.572 8.342 60.800 278.029 400.116 -> 112.725 -> 901.799 MByte/s p28 random-cyc-1dim :( 27.439) 0.036 0.570 8.312 61.189 285.578 397.027 -> 113.327 -> 906.613 MByte/s p29 random-cyc-1dim :( 27.425) 0.036 0.570 8.283 61.098 275.784 394.470 -> 113.649 -> 909.195 MByte/s p30 random-cyc-1dim :( 27.393) 0.037 0.570 8.115 60.637 264.492 364.516 -> 105.894 -> 847.151 MByte/s p31 random-cyc-1dim :( 27.740) 0.036 0.563 8.065 59.548 220.519 286.947 -> 86.899 -> 695.191 MByte/s p32 random-cyc-1dim :( 27.750) 0.036 0.562 8.094 59.721 211.093 284.644 -> 86.216 -> 689.732 MByte/s p33 random-cyc-1dim :( 27.593) 0.036 0.567 8.179 61.080 281.546 402.776 -> 112.293 -> 898.343 MByte/s p34 random-cyc-1dim :( 27.496) 0.036 0.572 8.266 61.384 278.254 381.101 -> 111.269 -> 890.148 MByte/s p35 random-cyc-1dim :( 27.608) 0.036 0.554 8.099 58.157 221.360 288.249 -> 85.859 -> 686.874 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 24.902) 0.040 0.596 8.935 72.234 286.553 363.781 -> 111.868 -> 894.943 MByte/s p37 best bi-section :( 24.503) 0.020 0.303 4.549 44.269 207.421 232.307 -> 72.159 -> 577.270 MByte/s p38 worst bi-section :( 24.580) 0.020 0.302 4.538 44.051 208.559 234.732 -> 72.828 -> 582.628 MByte/s p39 one PingPong Pair :( 23.430) 0.005 0.080 1.145 9.887 59.239 96.987 -> 25.350 -> 202.798 MByte/s p40 acyclic-2dim-all :( 19.883) 0.031 0.470 6.783 57.156 284.613 339.098 -> 106.035 -> 848.279 MByte/s p41 acyclic-3dim-all :( 16.059) 0.031 0.471 6.769 61.879 322.207 413.639 -> 120.952 -> 967.618 MByte/s p42 cyclic-2dim-x :( 15.460) 0.065 1.020 14.523 129.498 400.593 517.048 -> 169.147 -> 1353.179 MByte/s p43 cyclic-2dim-y :( 28.137) 0.036 0.561 8.149 56.754 215.476 229.643 -> 76.686 -> 613.491 MByte/s p44 cyclic-2dim-all :( 19.548) 0.051 0.782 11.279 90.600 320.862 416.605 -> 126.495 -> 1011.960 MByte/s p45 cyclic-3dim-x :( 15.521) 0.064 1.015 14.576 126.299 416.378 519.580 -> 178.852 -> 1430.815 MByte/s p46 cyclic-3dim-y :( 15.687) 0.064 1.006 14.396 125.957 438.562 519.001 -> 182.894 -> 1463.154 MByte/s p47 cyclic-3dim-z :( 28.187) 0.035 0.561 8.182 58.960 213.104 221.956 -> 76.665 -> 613.320 MByte/s p48 cyclic-3dim-all :( 19.674) 0.051 0.784 11.336 87.935 328.682 401.113 -> 126.580 -> 1012.644 MByte/s log_avg of all rings : 0.035 0.545 7.896 57.578 189.220 230.015 || 73.361 -> 586.887 MByte/s log_avg of all random : 0.037 0.565 8.204 60.944 247.645 333.622 || 98.669 -> 789.352 MByte/s log_avg(ring,random) : 0.036 0.555 8.049 59.237 216.470 277.016 || 85.079 -> 680.633 MByte/s * size -> accumulated on all pr.: 0.287 4.439 64.391 473.896 1731.762 2216.132 || 680.633 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 680.633 MByte/s on 8 processes ( = 85.079 MByte/s * 8 processes) Ping-pong latency: 23.430 microsec Ping-pong bandwidth: 775.896 MByte/s at Lmax= 8.000 MByte (MByte/s=1e6 Byte/s) (MByte=2**20 Byte) system parameters : 8 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 16:26:50 1999 Total execution wall clock time = 88 seconds SECTION-BEFF-END b_eff = 680.633 MB/s = 85.079 * 8 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000