b_eff = 1287.352 MB/s = 107.279 * 12 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 = 12 2-dim-paterns: size = 4 * 3 3-dim-paterns: size = 3 * 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-6*2fix 1=ring-3*4fix 2=ring-1*12fix 3=ring-1*12fix 4=ring-1*12fix 5=ring-1*12fix 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: 94.838 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 7.9e-01 4.7e-03 3.7e-02 235 6.2e-01 3.7e-03 2.9e-02 237 6.2e-01 3.7e-03 2.9e-02 2 160 4.2e-01 2.5e-03 2.0e-02 157 4.2e-01 2.5e-03 2.0e-02 160 4.2e-01 2.5e-03 2.0e-02 4 159 4.2e-01 2.5e-03 2.0e-02 159 4.3e-01 2.5e-03 2.0e-02 158 4.2e-01 2.5e-03 2.0e-02 8 158 4.2e-01 2.5e-03 2.0e-02 158 4.2e-01 2.5e-03 2.0e-02 158 4.2e-01 2.5e-03 2.0e-02 16 157 4.4e-01 2.5e-03 2.1e-02 158 4.3e-01 2.6e-03 2.1e-02 158 4.3e-01 2.5e-03 2.1e-02 32 157 4.4e-01 2.6e-03 2.1e-02 153 4.3e-01 2.5e-03 2.1e-02 157 4.4e-01 2.5e-03 2.1e-02 64 153 4.3e-01 2.6e-03 2.1e-02 154 4.4e-01 2.6e-03 2.1e-02 156 4.4e-01 2.6e-03 2.1e-02 128 147 4.4e-01 2.7e-03 2.1e-02 148 4.4e-01 2.7e-03 2.2e-02 149 4.4e-01 2.7e-03 2.2e-02 256 138 4.1e-01 2.5e-03 2.0e-02 138 4.3e-01 2.5e-03 2.0e-02 137 4.1e-01 2.4e-03 2.0e-02 512 140 4.3e-01 2.6e-03 2.1e-02 135 4.3e-01 2.5e-03 2.0e-02 142 4.3e-01 2.6e-03 2.1e-02 1024 137 4.3e-01 2.7e-03 2.1e-02 136 4.3e-01 2.6e-03 2.1e-02 137 4.3e-01 2.8e-03 2.1e-02 2048 128 6.8e-01 3.3e-03 2.9e-02 132 6.9e-01 3.4e-03 3.0e-02 124 6.5e-01 3.2e-03 2.9e-02 4096 96 6.3e-01 3.2e-03 2.7e-02 96 6.2e-01 3.2e-03 2.7e-02 97 6.3e-01 3.1e-03 2.7e-02 10624 58 7.1e-01 2.9e-03 3.0e-02 57 7.0e-01 3.0e-03 2.8e-02 59 7.2e-01 3.1e-03 2.9e-02 27554 39 8.1e-01 3.3e-03 3.2e-02 36 7.5e-01 3.0e-03 2.9e-02 36 7.6e-01 2.8e-03 2.9e-02 71468 23 9.8e-01 3.6e-03 3.8e-02 23 9.9e-01 4.3e-03 3.9e-02 24 1.0e+00 4.1e-03 4.0e-02 185364 12 1.0e+00 4.9e-03 4.1e-02 10 8.5e-01 4.8e-03 3.4e-02 11 9.2e-01 4.5e-03 3.8e-02 480774 4 7.9e-01 4.2e-03 3.3e-02 3 5.9e-01 3.1e-03 2.5e-02 4 7.6e-01 4.1e-03 3.3e-02 1246974 1 4.7e-01 2.4e-03 2.0e-02 1 4.3e-01 2.3e-03 2.1e-02 1 4.2e-01 2.4e-03 2.1e-02 3234251 1 3.8e-01 6.1e-03 3.5e-02 M 1 5.6e-01 6.1e-03 2.9e-02 M 1 5.5e-01 6.1e-03 4.2e-02 M 8388608 1 8.8e-01 1.4e-02 8.9e-02 R 1 1.3e+00 1.4e-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 4.4e+00 8.2e-02 1.2e-01 27 4.0e-01 7.3e-03 9.0e-03 13 1.9e-01 3.5e-03 4.3e-03 2 150 2.2e+00 4.1e-02 4.9e-02 13 1.9e-01 3.6e-03 4.4e-03 9 1.3e-01 2.5e-03 3.0e-03 4 75 1.1e+00 2.0e-02 2.5e-02 9 1.3e-01 2.4e-03 3.0e-03 9 1.3e-01 2.4e-03 3.0e-03 8 37 5.4e-01 1.0e-02 1.2e-02 9 1.3e-01 2.4e-03 3.0e-03 9 1.3e-01 2.4e-03 3.0e-03 16 18 2.7e-01 5.0e-03 6.0e-03 9 1.3e-01 2.6e-03 3.0e-03 9 1.3e-01 2.5e-03 3.0e-03 32 9 1.3e-01 2.5e-03 3.0e-03 8 1.2e-01 2.2e-03 2.7e-03 9 1.3e-01 2.5e-03 3.1e-03 64 9 1.3e-01 2.5e-03 3.0e-03 9 1.3e-01 2.5e-03 3.1e-03 9 1.3e-01 2.5e-03 3.1e-03 128 8 1.2e-01 2.3e-03 2.8e-03 8 1.2e-01 2.2e-03 2.8e-03 9 1.3e-01 2.5e-03 3.1e-03 256 8 1.2e-01 2.2e-03 2.8e-03 9 1.4e-01 2.6e-03 3.2e-03 8 1.2e-01 2.2e-03 2.8e-03 512 8 1.2e-01 2.3e-03 2.9e-03 8 1.2e-01 2.3e-03 2.9e-03 8 1.2e-01 2.2e-03 2.9e-03 1024 8 1.2e-01 2.3e-03 2.8e-03 8 1.2e-01 2.3e-03 2.9e-03 8 1.2e-01 2.2e-03 2.9e-03 2048 8 1.4e-01 2.6e-03 3.7e-03 8 1.4e-01 2.6e-03 3.7e-03 8 1.5e-01 2.6e-03 3.6e-03 4096 7 1.4e-01 2.4e-03 3.6e-03 7 1.4e-01 2.3e-03 3.7e-03 7 1.4e-01 2.3e-03 3.6e-03 10624 5 1.4e-01 1.8e-03 4.6e-03 5 1.4e-01 1.7e-03 3.8e-03 5 1.4e-01 1.7e-03 3.6e-03 27554 5 2.0e-01 2.1e-03 5.8e-03 5 2.0e-01 2.1e-03 5.9e-03 5 2.0e-01 2.1e-03 5.6e-03 71468 4 2.8e-01 2.8e-03 8.2e-03 4 2.7e-01 2.6e-03 8.2e-03 4 2.7e-01 2.7e-03 9.9e-03 185364 2 2.9e-01 2.7e-03 1.0e-02 2 2.8e-01 2.7e-03 8.4e-03 2 2.8e-01 2.6e-03 8.9e-03 480774 1 3.3e-01 3.0e-03 1.1e-02 1 3.3e-01 2.9e-03 1.0e-02 1 3.2e-01 2.8e-03 9.9e-03 1246974 1 6.8e-01 6.4e-03 2.1e-02 1 6.9e-01 6.5e-03 2.2e-02 1 7.0e-01 6.4e-03 2.6e-02 3234251 1 5.0e-02 2.1e-02 2.8e-02 M 1 2.1e-02 2.1e-02 2.1e-02 M 1 2.1e-02 2.1e-02 2.1e-02 M 8388608 1 1.3e-01 5.4e-02 7.2e-02 R 1 5.3e-02 5.3e-02 5.3e-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=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 1.4e+00 1.1e-02 5.1e-02 103 4.7e-01 3.7e-03 1.7e-02 105 4.7e-01 3.7e-03 1.8e-02 2 150 6.9e-01 5.4e-03 2.6e-02 70 3.2e-01 2.5e-03 1.2e-02 71 3.2e-01 2.5e-03 1.2e-02 4 75 3.5e-01 2.7e-03 1.3e-02 70 3.2e-01 2.5e-03 1.2e-02 69 3.1e-01 2.5e-03 1.2e-02 8 68 3.1e-01 2.5e-03 1.2e-02 69 3.1e-01 2.4e-03 1.2e-02 70 3.2e-01 2.5e-03 1.2e-02 16 69 3.3e-01 2.5e-03 1.2e-02 70 3.3e-01 2.5e-03 1.2e-02 69 3.2e-01 2.5e-03 1.2e-02 32 68 3.3e-01 2.6e-03 1.2e-02 69 3.3e-01 2.5e-03 1.2e-02 69 3.2e-01 2.5e-03 1.2e-02 64 66 3.2e-01 2.5e-03 1.2e-02 69 3.3e-01 2.6e-03 1.2e-02 69 3.3e-01 2.5e-03 1.2e-02 128 65 3.2e-01 2.5e-03 1.2e-02 67 3.3e-01 2.6e-03 1.2e-02 68 3.3e-01 2.6e-03 1.2e-02 256 64 3.2e-01 2.5e-03 1.2e-02 65 3.3e-01 2.6e-03 1.3e-02 65 3.2e-01 2.5e-03 1.2e-02 512 64 3.3e-01 2.5e-03 1.2e-02 63 3.3e-01 2.6e-03 1.2e-02 64 3.2e-01 2.4e-03 1.2e-02 1024 63 3.3e-01 2.5e-03 1.2e-02 61 3.1e-01 2.4e-03 1.2e-02 65 3.3e-01 2.5e-03 1.2e-02 2048 61 4.2e-01 2.8e-03 1.5e-02 63 4.3e-01 2.9e-03 1.6e-02 63 4.2e-01 2.8e-03 1.6e-02 4096 55 4.5e-01 2.9e-03 1.6e-02 55 4.5e-01 2.8e-03 1.6e-02 55 4.5e-01 2.9e-03 1.6e-02 10624 36 4.3e-01 2.7e-03 1.8e-02 37 4.3e-01 2.8e-03 1.7e-02 36 4.2e-01 2.8e-03 1.5e-02 27554 25 4.7e-01 2.5e-03 1.8e-02 25 4.6e-01 2.8e-03 1.6e-02 25 4.5e-01 2.8e-03 1.6e-02 71468 19 7.8e-01 3.6e-03 3.2e-02 17 6.8e-01 3.5e-03 2.8e-02 16 6.4e-01 3.1e-03 2.4e-02 185364 10 7.8e-01 4.2e-03 2.8e-02 9 6.9e-01 3.9e-03 2.6e-02 9 6.9e-01 3.8e-03 2.7e-02 480774 4 7.4e-01 4.0e-03 3.1e-02 4 7.2e-01 4.1e-03 2.9e-02 4 7.4e-01 4.4e-03 3.1e-02 1246974 1 4.2e-01 2.2e-03 1.7e-02 1 4.3e-01 2.3e-03 1.7e-02 1 4.3e-01 2.2e-03 2.3e-02 3234251 1 7.0e-01 6.2e-03 5.3e-02 M 1 4.9e-01 6.5e-03 5.5e-02 M 1 4.7e-01 6.3e-03 4.1e-02 M 8388608 1 1.6e+00 1.7e-02 1.2e-01 R 1 1.1e+00 1.6e-02 1.2e-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 = 94.838 sec sum of max elapsed time per entries above = 94.136 sec difference to elapsed time = 0.702 sec = 0.7% sum based on fastest repetition = 85.698 sec difference to elapsed time = 9.140 sec = 9.6% 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-6*2fix 1 12 1.00 1.00 0 ( 2 2 2 ) p01 ring-3*4fix 2 24 2.00 1.00 0 ( 2 0 0 ) p02 ring-1*12fix 2 24 2.00 1.00 0 ( 2 2 0 ) p03 ring-1*12fix 2 24 2.00 1.00 0 ( 2 0 2 ) p04 ring-1*12fix 2 24 2.00 1.00 0 ( 0 0 0 ) p05 ring-1*12fix 2 24 2.00 1.00 0 ( 2 0 0 ) p06 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 2 0 ) p07 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 2 2 ) p08 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 0 ) p09 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 2 2 ) p10 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 2 0 ) p11 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 2 2 ) p12 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 0 0 ) p13 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 0 ) p14 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 0 ) p15 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 0 0 ) p16 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 2 2 ) p17 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 2 ) p18 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 0 0 ) p19 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 0 ) p20 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 0 0 ) p21 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 0 ) p22 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 0 0 ) p23 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 2 2 ) p24 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 2 2 ) p25 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 0 2 ) p26 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 2 2 ) p27 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 0 ) p28 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 0 ) p29 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 0 ) p30 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 0 2 ) p31 random-cyc-1dim 2 24 2.00 1.00 0 ( 2 2 2 ) p32 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 0 2 ) p33 random-cyc-1dim 2 24 2.00 1.00 0 ( 1 0 0 ) p34 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 0 0 ) p35 random-cyc-1dim 2 24 2.00 1.00 0 ( 0 2 2 ) p36 worst-cyc-1dim 2 24 2.00 1.00 0 ( 2 2 0 ) p37 best bi-section 2 12 1.00 0.50 0 ( 2 0 2 ) p38 worst bi-section 2 12 1.00 0.50 0 ( 1 1 1 ) p39 one PingPong Pair 2 2 1.00 0.50 10 ( 0 0 0 ) p40 acyclic-2dim-all 4 34 2.83 0.71 0 ( 2 2 2 ) p41 acyclic-3dim-all 6 40 3.33 0.56 0 ( 2 2 2 ) p42 cyclic-2dim-x 2 24 2.00 1.00 0 ( 0 0 0 ) p43 cyclic-2dim-y 2 24 2.00 1.00 0 ( 0 0 0 ) p44 cyclic-2dim-all 4 48 4.00 1.00 0 ( 2 2 2 ) p45 cyclic-3dim-x 2 24 2.00 1.00 0 ( 0 2 0 ) p46 cyclic-3dim-y 1 12 1.00 1.00 0 ( 2 2 2 ) p47 cyclic-3dim-z 1 12 1.00 1.00 0 ( 2 2 2 ) p48 cyclic-3dim-all 4 48 4.00 1.00 0 ( 2 2 2 ) SECTION-PATTERN-END Printing accumulated bandwidth / number of processes = bandwidth per process (in MBytes/sec = 1e6 Bytes/sec) Additionally the accumulated value processes is printed in the last column Only the cyclic and random patterns are used to compute b_eff (except p00), the other patterns are only as information. SECTION-BY-METHODS-BEGIN 2nd analysis path, only as information, last row: for all mthd: logavg_pat ( avg_L ( max_rep ( b(L) ) ) ) || logavg_pat ( max_mthd ( avg_L ( max_rep ( b(L) ) ) ) ) pattern / methods: Sndrcv Alltoal non-blk -> best method -> accumulated p00 ring-6*2fix : 167.426 81.500 151.733 -> 167.426 -> 2009.112 MByte/s p01 ring-3*4fix : 123.309 79.833 118.278 -> 123.309 -> 1479.703 MByte/s p02 ring-1*12fix : 122.597 75.082 120.924 -> 122.597 -> 1471.166 MByte/s p03 ring-1*12fix : 121.705 73.292 117.973 -> 121.705 -> 1460.465 MByte/s p04 ring-1*12fix : 125.191 76.259 120.688 -> 125.191 -> 1502.297 MByte/s p05 ring-1*12fix : 126.195 75.052 120.971 -> 126.195 -> 1514.343 MByte/s p06 random-cyc-1dim : 103.944 65.901 94.826 -> 103.944 -> 1247.333 MByte/s p07 random-cyc-1dim : 93.588 62.909 92.095 -> 93.588 -> 1123.058 MByte/s p08 random-cyc-1dim : 83.719 55.775 79.092 -> 83.719 -> 1004.623 MByte/s p09 random-cyc-1dim : 57.825 43.127 54.241 -> 57.825 -> 693.902 MByte/s p10 random-cyc-1dim : 101.559 65.837 96.481 -> 101.559 -> 1218.704 MByte/s p11 random-cyc-1dim : 81.752 55.151 80.191 -> 81.752 -> 981.024 MByte/s p12 random-cyc-1dim : 68.104 49.181 68.054 -> 68.104 -> 817.244 MByte/s p13 random-cyc-1dim : 86.317 57.291 79.447 -> 86.317 -> 1035.810 MByte/s p14 random-cyc-1dim : 82.549 56.640 81.102 -> 82.549 -> 990.589 MByte/s p15 random-cyc-1dim : 81.346 57.424 78.682 -> 81.346 -> 976.156 MByte/s p16 random-cyc-1dim : 81.968 55.441 82.671 -> 82.671 -> 992.047 MByte/s p17 random-cyc-1dim : 69.080 49.173 67.317 -> 69.080 -> 828.959 MByte/s p18 random-cyc-1dim : 68.783 49.925 66.078 -> 68.783 -> 825.395 MByte/s p19 random-cyc-1dim : 81.784 56.871 80.088 -> 81.784 -> 981.406 MByte/s p20 random-cyc-1dim : 82.278 57.356 81.506 -> 82.278 -> 987.335 MByte/s p21 random-cyc-1dim : 81.504 56.402 81.397 -> 81.504 -> 978.043 MByte/s p22 random-cyc-1dim : 108.019 67.275 101.976 -> 108.019 -> 1296.234 MByte/s p23 random-cyc-1dim : 80.787 56.159 80.882 -> 80.882 -> 970.579 MByte/s p24 random-cyc-1dim : 96.707 64.853 96.210 -> 96.707 -> 1160.486 MByte/s p25 random-cyc-1dim : 104.755 67.202 97.520 -> 104.755 -> 1257.061 MByte/s p26 random-cyc-1dim : 121.522 74.202 119.238 -> 121.522 -> 1458.260 MByte/s p27 random-cyc-1dim : 125.235 75.946 119.056 -> 125.235 -> 1502.824 MByte/s p28 random-cyc-1dim : 82.479 56.495 81.717 -> 82.479 -> 989.746 MByte/s p29 random-cyc-1dim : 79.779 55.615 78.347 -> 79.779 -> 957.350 MByte/s p30 random-cyc-1dim : 106.164 67.204 100.796 -> 106.164 -> 1273.968 MByte/s p31 random-cyc-1dim : 84.822 54.668 79.854 -> 84.822 -> 1017.865 MByte/s p32 random-cyc-1dim : 82.170 57.946 81.534 -> 82.170 -> 986.039 MByte/s p33 random-cyc-1dim : 82.262 57.862 81.300 -> 82.262 -> 987.142 MByte/s p34 random-cyc-1dim : 78.586 56.521 78.489 -> 78.586 -> 943.031 MByte/s p35 random-cyc-1dim : 68.621 49.250 69.495 -> 69.495 -> 833.943 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 51.547 39.973 48.132 -> 51.547 -> 618.569 MByte/s p37 best bi-section : 148.396 83.575 159.149 -> 159.149 -> 1909.784 MByte/s p38 worst bi-section : 42.876 38.443 50.937 -> 50.937 -> 611.247 MByte/s p39 one PingPong Pair : 27.303 9.009 9.009 -> 27.303 -> 327.640 MByte/s p40 acyclic-2dim-all : 92.075 78.079 101.804 -> 101.804 -> 1221.650 MByte/s p41 acyclic-3dim-all : 79.440 75.981 96.087 -> 96.087 -> 1153.047 MByte/s p42 cyclic-2dim-x : 90.505 55.939 85.413 -> 90.505 -> 1086.054 MByte/s p43 cyclic-2dim-y : 159.492 83.569 142.926 -> 159.492 -> 1913.902 MByte/s p44 cyclic-2dim-all : 112.452 81.448 116.336 -> 116.336 -> 1396.037 MByte/s p45 cyclic-3dim-x : 73.101 47.881 67.771 -> 73.101 -> 877.210 MByte/s p46 cyclic-3dim-y : 106.321 61.485 91.484 -> 106.321 -> 1275.854 MByte/s p47 cyclic-3dim-z : 168.685 81.212 149.231 -> 168.685 -> 2024.217 MByte/s p48 cyclic-3dim-all : 91.126 74.927 94.181 -> 94.181 -> 1130.169 MByte/s log_avg of all rings : 130.178 76.783 124.577 || 130.178 -> 1562.135 MByte/s log_avg of all random : 85.647 58.074 83.189 || 85.711 -> 1028.532 MByte/s log_avg(ring,random) : 105.591 66.776 101.801 ||(105.630 -> 1267.559)MByte/s * size -> accumulated on all pr.: 1267.087 801.316 1221.613 ||(1267.559)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-6*2fix : 168.175 162.922 170.034 -> 170.034 -> 2040.413 MByte/s p01 ring-3*4fix : 114.888 122.035 122.420 -> 122.420 -> 1469.039 MByte/s p02 ring-1*12fix : 110.376 112.414 123.560 -> 123.560 -> 1482.722 MByte/s p03 ring-1*12fix : 111.902 119.714 116.224 -> 119.714 -> 1436.571 MByte/s p04 ring-1*12fix : 117.640 119.767 124.778 -> 124.778 -> 1497.334 MByte/s p05 ring-1*12fix : 123.067 116.646 118.643 -> 123.067 -> 1476.807 MByte/s p06 random-cyc-1dim : 87.035 101.437 103.257 -> 103.257 -> 1239.083 MByte/s p07 random-cyc-1dim : 84.804 94.221 96.853 -> 96.853 -> 1162.234 MByte/s p08 random-cyc-1dim : 71.096 82.580 84.283 -> 84.283 -> 1011.397 MByte/s p09 random-cyc-1dim : 51.683 57.524 57.230 -> 57.524 -> 690.284 MByte/s p10 random-cyc-1dim : 85.654 96.786 100.963 -> 100.963 -> 1211.560 MByte/s p11 random-cyc-1dim : 75.459 81.823 82.740 -> 82.740 -> 992.877 MByte/s p12 random-cyc-1dim : 63.253 69.657 69.269 -> 69.657 -> 835.890 MByte/s p13 random-cyc-1dim : 74.309 86.698 82.941 -> 86.698 -> 1040.370 MByte/s p14 random-cyc-1dim : 71.313 83.550 83.420 -> 83.550 -> 1002.603 MByte/s p15 random-cyc-1dim : 78.057 81.749 82.530 -> 82.530 -> 990.364 MByte/s p16 random-cyc-1dim : 79.923 83.406 84.438 -> 84.438 -> 1013.260 MByte/s p17 random-cyc-1dim : 64.764 69.517 66.625 -> 69.517 -> 834.203 MByte/s p18 random-cyc-1dim : 63.212 69.360 67.341 -> 69.360 -> 832.320 MByte/s p19 random-cyc-1dim : 77.243 84.205 80.712 -> 84.205 -> 1010.458 MByte/s p20 random-cyc-1dim : 81.437 81.947 83.180 -> 83.180 -> 998.156 MByte/s p21 random-cyc-1dim : 78.214 82.471 82.122 -> 82.471 -> 989.654 MByte/s p22 random-cyc-1dim : 100.717 103.717 106.579 -> 106.579 -> 1278.950 MByte/s p23 random-cyc-1dim : 73.686 81.013 74.983 -> 81.013 -> 972.158 MByte/s p24 random-cyc-1dim : 87.838 95.486 98.830 -> 98.830 -> 1185.956 MByte/s p25 random-cyc-1dim : 100.879 100.893 102.095 -> 102.095 -> 1225.143 MByte/s p26 random-cyc-1dim : 118.349 117.600 117.327 -> 118.349 -> 1420.187 MByte/s p27 random-cyc-1dim : 117.861 121.542 123.783 -> 123.783 -> 1485.394 MByte/s p28 random-cyc-1dim : 81.061 83.132 82.326 -> 83.132 -> 997.590 MByte/s p29 random-cyc-1dim : 76.548 80.584 79.116 -> 80.584 -> 967.014 MByte/s p30 random-cyc-1dim : 94.738 102.918 102.204 -> 102.918 -> 1235.017 MByte/s p31 random-cyc-1dim : 82.101 82.997 82.240 -> 82.997 -> 995.966 MByte/s p32 random-cyc-1dim : 79.614 83.434 81.908 -> 83.434 -> 1001.205 MByte/s p33 random-cyc-1dim : 70.440 82.564 77.560 -> 82.564 -> 990.766 MByte/s p34 random-cyc-1dim : 78.302 80.149 80.884 -> 80.884 -> 970.614 MByte/s p35 random-cyc-1dim : 71.070 68.364 69.712 -> 71.070 -> 852.837 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 49.896 49.647 51.700 -> 51.700 -> 620.401 MByte/s p37 best bi-section : 141.499 149.815 146.509 -> 149.815 -> 1797.782 MByte/s p38 worst bi-section : 50.754 51.149 50.827 -> 51.149 -> 613.784 MByte/s p39 one PingPong Pair : 26.519 27.211 26.942 -> 27.211 -> 326.529 MByte/s p40 acyclic-2dim-all : 97.184 100.332 101.352 -> 101.352 -> 1216.220 MByte/s p41 acyclic-3dim-all : 95.853 91.541 92.430 -> 95.853 -> 1150.241 MByte/s p42 cyclic-2dim-x : 88.689 87.844 90.008 -> 90.008 -> 1080.096 MByte/s p43 cyclic-2dim-y : 146.850 153.744 154.208 -> 154.208 -> 1850.499 MByte/s p44 cyclic-2dim-all : 115.261 116.221 114.281 -> 116.221 -> 1394.657 MByte/s p45 cyclic-3dim-x : 72.513 70.426 72.550 -> 72.550 -> 870.596 MByte/s p46 cyclic-3dim-y : 102.976 101.410 102.084 -> 102.976 -> 1235.716 MByte/s p47 cyclic-3dim-z : 163.802 149.870 165.230 -> 165.230 -> 1982.755 MByte/s p48 cyclic-3dim-all : 93.606 93.072 92.528 -> 93.606 -> 1123.276 MByte/s log_avg of all rings : 122.954 124.577 128.132 || 129.553 -> 1554.641 MByte/s log_avg of all random : 79.485 85.300 85.005 || 86.163 -> 1033.957 MByte/s log_avg(ring,random) : 98.859 103.085 104.364 ||(105.654 -> 1267.845)MByte/s * size -> accumulated on all pr.: 1186.303 1237.017 1252.372 ||(1267.845)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-6*2fix p00 method 0 =Sndrcv :( 15.974) 0.063 1.003 13.668 118.478 389.117 496.723 -> 167.426 -> 2009.112 MByte/s p00 method 1 =Alltoal :(322.037) 0.003 0.051 0.805 12.493 132.828 496.723 -> 81.500 -> 977.997 MByte/s p00 method 2 =non-blk :( 37.133) 0.027 0.432 6.521 72.931 369.987 496.723 -> 151.733 -> 1820.797 MByte/s p01 ring-3*4fix p01 method 0 =Sndrcv :( 25.238) 0.040 0.590 8.667 69.614 302.908 412.642 -> 123.309 -> 1479.703 MByte/s p01 method 1 =Alltoal :(159.620) 0.006 0.100 1.579 21.954 173.403 412.642 -> 79.833 -> 957.998 MByte/s p01 method 2 =non-blk :( 47.172) 0.021 0.328 4.955 52.016 306.105 412.642 -> 118.278 -> 1419.335 MByte/s p02 ring-1*12fix p02 method 0 =Sndrcv :( 25.027) 0.040 0.596 8.778 69.806 285.076 433.643 -> 122.597 -> 1471.166 MByte/s p02 method 1 =Alltoal :(158.269) 0.006 0.100 1.586 22.479 131.277 433.643 -> 75.082 -> 900.985 MByte/s p02 method 2 =non-blk :( 44.486) 0.022 0.348 5.187 53.132 286.424 433.643 -> 120.924 -> 1451.085 MByte/s p03 ring-1*12fix p03 method 0 =Sndrcv :( 25.015) 0.040 0.595 8.805 69.479 288.752 432.984 -> 121.705 -> 1460.465 MByte/s p03 method 1 =Alltoal :(159.893) 0.006 0.101 1.583 22.603 128.458 432.984 -> 73.292 -> 879.503 MByte/s p03 method 2 =non-blk :( 44.371) 0.023 0.348 5.239 53.466 280.218 432.984 -> 117.973 -> 1415.674 MByte/s p04 ring-1*12fix p04 method 0 =Sndrcv :( 25.051) 0.040 0.594 8.822 70.305 293.957 441.111 -> 125.191 -> 1502.297 MByte/s p04 method 1 =Alltoal :(160.010) 0.006 0.101 1.571 22.470 131.792 441.111 -> 76.259 -> 915.110 MByte/s p04 method 2 =non-blk :( 44.400) 0.023 0.344 5.282 53.188 296.900 441.111 -> 120.688 -> 1448.257 MByte/s p05 ring-1*12fix p05 method 0 =Sndrcv :( 25.164) 0.040 0.596 8.892 69.781 304.170 425.818 -> 126.195 -> 1514.343 MByte/s p05 method 1 =Alltoal :(158.461) 0.006 0.100 1.580 22.567 132.002 425.818 -> 75.052 -> 900.621 MByte/s p05 method 2 =non-blk :( 44.738) 0.022 0.343 5.235 53.779 290.289 425.818 -> 120.971 -> 1451.656 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 27.878) 0.036 0.561 8.183 57.690 253.592 351.282 -> 103.944 -> 1247.333 MByte/s p06 method 1 =Alltoal :(153.852) 0.006 0.104 1.638 20.657 124.341 351.282 -> 65.901 -> 790.812 MByte/s p06 method 2 =non-blk :( 47.243) 0.021 0.324 4.900 46.254 253.020 351.282 -> 94.826 -> 1137.916 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 27.901) 0.036 0.558 8.158 57.919 213.575 335.739 -> 93.588 -> 1123.058 MByte/s p07 method 1 =Alltoal :(154.595) 0.006 0.104 1.634 21.176 128.190 335.739 -> 62.909 -> 754.910 MByte/s p07 method 2 =non-blk :( 47.486) 0.021 0.326 4.807 45.151 240.646 335.739 -> 92.095 -> 1105.144 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 27.896) 0.036 0.551 8.047 57.316 208.957 280.340 -> 83.719 -> 1004.623 MByte/s p08 method 1 =Alltoal :(144.615) 0.007 0.110 1.704 19.794 113.165 280.340 -> 55.775 -> 669.298 MByte/s p08 method 2 =non-blk :( 47.443) 0.021 0.325 4.797 45.580 211.080 280.340 -> 79.092 -> 949.110 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 28.496) 0.035 0.532 7.824 55.440 144.723 174.835 -> 57.825 -> 693.902 MByte/s p09 method 1 =Alltoal :(135.729) 0.007 0.116 1.811 17.403 107.818 174.835 -> 43.127 -> 517.526 MByte/s p09 method 2 =non-blk :( 49.814) 0.020 0.308 4.643 43.545 145.092 174.835 -> 54.241 -> 650.897 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 27.784) 0.036 0.558 8.236 57.303 263.602 347.556 -> 101.559 -> 1218.704 MByte/s p10 method 1 =Alltoal :(153.922) 0.006 0.104 1.627 21.160 128.346 347.556 -> 65.837 -> 790.048 MByte/s p10 method 2 =non-blk :( 47.823) 0.021 0.325 4.811 45.789 244.957 347.556 -> 96.481 -> 1157.768 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 27.918) 0.036 0.550 8.034 57.341 195.214 273.570 -> 81.752 -> 981.024 MByte/s p11 method 1 =Alltoal :(145.322) 0.007 0.110 1.700 19.699 116.143 273.570 -> 55.151 -> 661.816 MByte/s p11 method 2 =non-blk :( 47.586) 0.021 0.327 4.867 45.119 206.176 273.570 -> 80.191 -> 962.297 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 28.297) 0.035 0.542 8.026 55.537 172.155 213.635 -> 68.104 -> 817.244 MByte/s p12 method 1 =Alltoal :(143.248) 0.007 0.112 1.753 18.344 113.581 213.635 -> 49.181 -> 590.173 MByte/s p12 method 2 =non-blk :( 49.048) 0.020 0.313 4.755 44.544 173.897 213.635 -> 68.054 -> 816.649 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 27.897) 0.036 0.557 8.213 57.576 217.819 297.020 -> 86.317 -> 1035.810 MByte/s p13 method 1 =Alltoal :(145.963) 0.007 0.109 1.688 19.511 122.676 297.020 -> 57.291 -> 687.491 MByte/s p13 method 2 =non-blk :( 47.600) 0.021 0.323 4.903 45.845 206.432 297.020 -> 79.447 -> 953.362 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 28.173) 0.035 0.555 8.136 57.133 193.877 285.521 -> 82.549 -> 990.589 MByte/s p14 method 1 =Alltoal :(146.504) 0.007 0.108 1.682 19.465 121.430 285.521 -> 56.640 -> 679.677 MByte/s p14 method 2 =non-blk :( 47.843) 0.021 0.324 4.846 45.502 209.596 285.521 -> 81.102 -> 973.221 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 28.057) 0.036 0.548 8.033 56.725 187.445 285.128 -> 81.346 -> 976.156 MByte/s p15 method 1 =Alltoal :(146.151) 0.007 0.108 1.689 19.340 124.717 285.128 -> 57.424 -> 689.083 MByte/s p15 method 2 =non-blk :( 48.990) 0.020 0.313 4.712 45.183 210.654 285.128 -> 78.682 -> 944.189 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 27.952) 0.036 0.550 8.078 56.917 186.475 275.737 -> 81.968 -> 983.610 MByte/s p16 method 1 =Alltoal :(145.487) 0.007 0.107 1.697 19.712 119.032 275.737 -> 55.441 -> 665.292 MByte/s p16 method 2 =non-blk :( 47.825) 0.021 0.322 4.797 44.962 221.978 275.737 -> 82.671 -> 992.047 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 28.184) 0.035 0.544 7.949 55.803 167.107 227.512 -> 69.080 -> 828.959 MByte/s p17 method 1 =Alltoal :(142.122) 0.007 0.111 1.760 18.216 110.386 227.512 -> 49.173 -> 590.078 MByte/s p17 method 2 =non-blk :( 49.243) 0.020 0.312 4.665 44.047 168.521 227.512 -> 67.317 -> 807.801 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 28.267) 0.035 0.543 7.918 56.489 162.477 227.232 -> 68.783 -> 825.395 MByte/s p18 method 1 =Alltoal :(142.193) 0.007 0.113 1.757 18.245 112.598 227.232 -> 49.925 -> 599.104 MByte/s p18 method 2 =non-blk :( 49.367) 0.020 0.309 4.664 44.739 173.969 227.232 -> 66.078 -> 792.936 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 27.903) 0.036 0.553 8.016 56.676 187.663 294.631 -> 81.784 -> 981.406 MByte/s p19 method 1 =Alltoal :(143.849) 0.007 0.108 1.712 19.545 117.561 294.631 -> 56.871 -> 682.451 MByte/s p19 method 2 =non-blk :( 48.005) 0.021 0.323 4.792 45.429 218.088 294.631 -> 80.088 -> 961.053 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 28.165) 0.036 0.549 8.033 56.364 191.731 290.264 -> 82.278 -> 987.335 MByte/s p20 method 1 =Alltoal :(146.614) 0.007 0.108 1.697 19.001 123.494 290.264 -> 57.356 -> 688.274 MByte/s p20 method 2 =non-blk :( 48.238) 0.021 0.319 4.737 45.052 212.519 290.264 -> 81.506 -> 978.076 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 27.844) 0.036 0.554 8.130 57.408 191.549 282.212 -> 81.504 -> 978.043 MByte/s p21 method 1 =Alltoal :(145.787) 0.007 0.109 1.704 19.445 121.293 282.212 -> 56.402 -> 676.827 MByte/s p21 method 2 =non-blk :( 48.129) 0.021 0.326 4.797 44.836 212.658 282.212 -> 81.397 -> 976.765 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 25.380) 0.039 0.590 8.754 67.667 268.873 366.972 -> 108.019 -> 1296.234 MByte/s p22 method 1 =Alltoal :(151.067) 0.007 0.106 1.642 20.860 129.784 366.972 -> 67.275 -> 807.305 MByte/s p22 method 2 =non-blk :( 45.791) 0.022 0.336 5.170 51.788 257.709 366.972 -> 101.976 -> 1223.714 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 27.907) 0.036 0.551 8.136 57.120 192.277 272.118 -> 80.787 -> 969.444 MByte/s p23 method 1 =Alltoal :(145.114) 0.007 0.110 1.655 19.585 121.352 272.118 -> 56.159 -> 673.911 MByte/s p23 method 2 =non-blk :( 47.962) 0.021 0.322 4.817 45.214 212.294 272.118 -> 80.882 -> 970.579 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 27.828) 0.036 0.557 8.278 58.224 215.154 338.982 -> 96.707 -> 1160.486 MByte/s p24 method 1 =Alltoal :(152.849) 0.007 0.104 1.622 21.005 131.719 338.982 -> 64.853 -> 778.233 MByte/s p24 method 2 =non-blk :( 47.452) 0.021 0.321 4.876 46.183 243.775 338.982 -> 96.210 -> 1154.515 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 27.823) 0.036 0.562 8.237 58.619 260.961 391.781 -> 104.755 -> 1257.061 MByte/s p25 method 1 =Alltoal :(153.813) 0.007 0.104 1.624 20.546 129.151 391.781 -> 67.202 -> 806.423 MByte/s p25 method 2 =non-blk :( 47.348) 0.021 0.323 4.886 46.536 232.170 391.781 -> 97.520 -> 1170.245 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 24.892) 0.040 0.595 8.824 69.565 297.367 432.715 -> 121.522 -> 1458.260 MByte/s p26 method 1 =Alltoal :(159.773) 0.006 0.101 1.578 22.278 136.321 432.715 -> 74.202 -> 890.427 MByte/s p26 method 2 =non-blk :( 44.443) 0.023 0.342 5.151 52.703 286.104 432.715 -> 119.238 -> 1430.851 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 24.985) 0.040 0.594 8.802 70.418 300.429 435.998 -> 125.235 -> 1502.824 MByte/s p27 method 1 =Alltoal :(158.503) 0.006 0.101 1.572 22.400 138.021 435.998 -> 75.946 -> 911.347 MByte/s p27 method 2 =non-blk :( 44.633) 0.022 0.344 5.248 53.321 303.958 435.998 -> 119.056 -> 1428.667 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 27.912) 0.036 0.555 8.177 56.840 190.292 285.395 -> 82.479 -> 989.746 MByte/s p28 method 1 =Alltoal :(148.588) 0.007 0.108 1.699 19.632 117.523 285.395 -> 56.495 -> 677.945 MByte/s p28 method 2 =non-blk :( 48.323) 0.021 0.315 4.904 45.169 206.049 285.395 -> 81.717 -> 980.603 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 28.017) 0.036 0.553 8.035 57.258 206.256 276.164 -> 79.779 -> 957.350 MByte/s p29 method 1 =Alltoal :(147.615) 0.007 0.108 1.696 19.511 120.876 276.164 -> 55.615 -> 667.378 MByte/s p29 method 2 =non-blk :( 48.219) 0.021 0.314 4.843 44.460 207.191 276.164 -> 78.347 -> 940.166 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 27.681) 0.036 0.564 8.255 57.496 265.074 382.230 -> 106.164 -> 1273.968 MByte/s p30 method 1 =Alltoal :(153.775) 0.007 0.104 1.582 20.495 124.635 382.230 -> 67.204 -> 806.443 MByte/s p30 method 2 =non-blk :( 47.605) 0.021 0.324 4.832 45.547 262.082 382.230 -> 100.796 -> 1209.555 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 27.850) 0.036 0.556 8.167 56.905 224.197 271.256 -> 84.822 -> 1017.865 MByte/s p31 method 1 =Alltoal :(145.843) 0.007 0.108 1.702 19.373 122.739 271.256 -> 54.668 -> 656.014 MByte/s p31 method 2 =non-blk :( 48.015) 0.021 0.321 4.873 45.723 207.626 271.256 -> 79.854 -> 958.248 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 28.042) 0.036 0.555 8.110 57.170 197.177 291.651 -> 82.170 -> 986.039 MByte/s p32 method 1 =Alltoal :(146.384) 0.007 0.107 1.689 19.518 118.047 291.651 -> 57.946 -> 695.349 MByte/s p32 method 2 =non-blk :( 48.015) 0.021 0.318 4.764 44.881 209.281 291.651 -> 81.534 -> 978.405 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 27.922) 0.036 0.556 8.097 56.553 186.398 289.502 -> 82.262 -> 987.142 MByte/s p33 method 1 =Alltoal :(148.058) 0.007 0.106 1.696 19.321 125.820 289.502 -> 57.862 -> 694.341 MByte/s p33 method 2 =non-blk :( 48.553) 0.021 0.318 4.774 45.007 209.344 289.502 -> 81.300 -> 975.598 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 27.934) 0.036 0.552 8.042 56.090 184.595 274.609 -> 78.586 -> 943.031 MByte/s p34 method 1 =Alltoal :(145.525) 0.007 0.111 1.710 19.578 121.770 274.609 -> 56.521 -> 678.249 MByte/s p34 method 2 =non-blk :( 48.081) 0.021 0.323 4.811 44.650 205.515 274.609 -> 78.489 -> 941.874 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 28.205) 0.035 0.538 7.928 56.533 156.244 223.908 -> 68.621 -> 823.452 MByte/s p35 method 1 =Alltoal :(142.905) 0.007 0.112 1.743 18.624 110.009 223.908 -> 49.250 -> 591.004 MByte/s p35 method 2 =non-blk :( 49.029) 0.020 0.312 4.651 44.473 177.911 223.908 -> 69.495 -> 833.943 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 29.209) 0.034 0.523 7.655 52.373 128.623 146.454 -> 51.547 -> 618.569 MByte/s p36 method 1 =Alltoal :(134.651) 0.007 0.114 1.841 16.708 103.281 146.454 -> 39.973 -> 479.679 MByte/s p36 method 2 =non-blk :( 51.578) 0.019 0.298 4.522 42.104 126.628 146.454 -> 48.132 -> 577.583 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 11.988) 0.042 0.635 9.122 90.916 332.242 579.725 -> 148.396 -> 1780.754 MByte/s p37 method 1 =Alltoal :(159.875) 0.003 0.051 0.798 12.407 139.370 579.725 -> 83.575 -> 1002.902 MByte/s p37 method 2 =non-blk :( 17.753) 0.028 0.436 6.625 76.238 384.043 579.725 -> 159.149 -> 1909.784 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 25.013) 0.020 0.299 4.422 35.393 120.398 157.031 -> 42.876 -> 514.517 MByte/s p38 method 1 =Alltoal :(160.272) 0.003 0.050 0.785 10.655 100.332 157.031 -> 38.443 -> 461.320 MByte/s p38 method 2 =non-blk :( 26.557) 0.019 0.285 4.520 42.642 141.923 157.031 -> 50.937 -> 611.247 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 11.527) 0.007 0.110 1.524 17.281 58.902 100.729 -> 27.303 -> 327.640 MByte/s p39 method 1 =Alltoal :(-------) 0.000 0.000 0.000 0.000 0.000 100.729 -> 9.009 -> 108.105 MByte/s p39 method 2 =non-blk :(-------) 0.000 0.000 0.000 0.000 0.000 100.729 -> 9.009 -> 108.105 MByte/s p40 acyclic-2dim-all p40 method 0 =Sndrcv :( 20.335) 0.035 0.525 7.795 62.231 217.236 342.065 -> 92.075 -> 1104.900 MByte/s p40 method 1 =Alltoal :( 80.347) 0.009 0.141 2.200 26.105 184.965 342.065 -> 78.079 -> 936.943 MByte/s p40 method 2 =non-blk :( 36.755) 0.019 0.300 4.509 48.010 265.847 342.065 -> 101.804 -> 1221.650 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 20.562) 0.027 0.406 5.963 48.984 179.794 315.179 -> 79.440 -> 953.282 MByte/s p41 method 1 =Alltoal :( 54.448) 0.010 0.159 2.436 26.832 188.235 315.179 -> 75.981 -> 911.767 MByte/s p41 method 2 =non-blk :( 26.076) 0.021 0.329 4.898 47.579 257.820 315.179 -> 96.087 -> 1153.047 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 25.561) 0.039 0.584 8.703 65.764 221.820 298.921 -> 90.505 -> 1086.054 MByte/s p42 method 1 =Alltoal :(160.212) 0.006 0.100 1.560 18.919 109.912 298.921 -> 55.939 -> 671.264 MByte/s p42 method 2 =non-blk :( 46.953) 0.021 0.326 5.003 51.545 223.613 298.921 -> 85.413 -> 1024.954 MByte/s p43 cyclic-2dim-y p43 method 0 =Sndrcv :( 16.101) 0.062 1.007 13.461 119.484 359.136 494.713 -> 159.492 -> 1913.902 MByte/s p43 method 1 =Alltoal :(159.668) 0.006 0.100 1.589 24.165 133.429 494.713 -> 83.569 -> 1002.833 MByte/s p43 method 2 =non-blk :( 35.143) 0.028 0.455 6.751 80.846 341.792 494.713 -> 142.926 -> 1715.110 MByte/s p44 cyclic-2dim-all p44 method 0 =Sndrcv :( 20.833) 0.048 0.716 10.546 83.445 276.921 374.579 -> 112.452 -> 1349.422 MByte/s p44 method 1 =Alltoal :( 81.172) 0.012 0.198 3.083 36.077 177.764 374.579 -> 81.448 -> 977.375 MByte/s p44 method 2 =non-blk :( 40.386) 0.025 0.385 5.867 62.348 308.194 374.579 -> 116.336 -> 1396.037 MByte/s p45 cyclic-3dim-x p45 method 0 =Sndrcv :( 28.057) 0.036 0.552 8.079 56.692 192.054 221.707 -> 73.101 -> 877.210 MByte/s p45 method 1 =Alltoal :(160.005) 0.006 0.099 1.543 16.881 107.038 221.707 -> 47.881 -> 574.572 MByte/s p45 method 2 =non-blk :( 48.195) 0.021 0.317 4.776 44.570 170.528 221.707 -> 67.771 -> 813.252 MByte/s p46 cyclic-3dim-y p46 method 0 =Sndrcv :( 28.009) 0.036 0.552 8.100 58.383 298.230 340.779 -> 106.321 -> 1275.854 MByte/s p46 method 1 =Alltoal :(318.833) 0.003 0.051 0.795 10.189 122.432 340.779 -> 61.485 -> 737.818 MByte/s p46 method 2 =non-blk :( 52.767) 0.019 0.283 4.411 45.101 224.293 340.779 -> 91.484 -> 1097.810 MByte/s p47 cyclic-3dim-z p47 method 0 =Sndrcv :( 16.156) 0.062 0.983 13.473 120.100 402.237 502.189 -> 168.685 -> 2024.217 MByte/s p47 method 1 =Alltoal :(315.538) 0.003 0.051 0.802 12.477 133.455 502.189 -> 81.212 -> 974.539 MByte/s p47 method 2 =non-blk :( 36.733) 0.027 0.439 6.496 75.700 331.466 502.189 -> 149.231 -> 1790.776 MByte/s p48 cyclic-3dim-all p48 method 0 =Sndrcv :( 23.848) 0.042 0.634 9.331 69.958 225.690 295.340 -> 91.126 -> 1093.508 MByte/s p48 method 1 =Alltoal :( 80.290) 0.012 0.197 3.035 31.991 190.263 295.340 -> 74.927 -> 899.121 MByte/s p48 method 2 =non-blk :( 42.076) 0.024 0.370 5.544 55.426 258.266 295.340 -> 94.181 -> 1130.169 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.043 0.648 9.464 76.231 308.825 439.721 || 130.178 -> 1562.135 MByte/s - ring, method 1 = Alltoal: 0.006 0.090 1.412 20.333 137.497 439.721 || 76.783 -> 921.394 MByte/s - ring, method 2 = non-blk: 0.023 0.356 5.381 55.995 303.622 439.721 || 124.577 -> 1494.922 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.036 0.556 8.159 58.094 205.903 293.218 || 85.647 -> 1027.766 MByte/s - random, method 1 = Alltoal: 0.007 0.108 1.681 19.750 121.533 293.218 || 58.074 -> 696.887 MByte/s - random, method 2 = non-blk: 0.021 0.322 4.834 45.853 215.146 293.218 || 83.189 -> 998.272 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.039 0.600 8.787 66.547 252.167 359.074 || 105.591 -> 1267.087 MByte/s - average, method 1 = Alltoal: 0.006 0.098 1.541 20.040 129.269 359.074 || 66.776 -> 801.316 MByte/s - average, method 2 = non-blk: 0.022 0.338 5.100 50.671 255.584 359.074 || 101.801 -> 1221.613 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 0.473 7.205 105.442 798.569 3026.001 4308.890 || 1267.087 MByte/s - accumulated, mthd 1 = Alltoal: 0.074 1.179 18.487 240.475 1551.224 4308.890 || 801.316 MByte/s - accumulated, mthd 2 = non-blk: 0.263 4.059 61.205 608.052 3067.010 4308.890 || 1221.613 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 0.473 0.039 0.043 0.036 0.039 0.006 0.022 2 0.942 0.079 0.086 0.072 0.079 0.012 0.044 4 1.865 0.155 0.170 0.142 0.155 0.025 0.086 8 3.777 0.315 0.344 0.288 0.315 0.049 0.175 16 7.205 0.600 0.648 0.556 0.600 0.098 0.338 32 14.170 1.181 1.271 1.097 1.181 0.196 0.668 64 27.853 2.321 2.500 2.155 2.321 0.390 1.326 128 53.412 4.451 4.762 4.160 4.451 0.777 2.584 256 105.442 8.787 9.464 8.159 8.787 1.541 5.100 512 207.468 17.289 18.704 15.981 17.289 3.069 10.035 1024 398.822 33.235 35.628 31.003 33.235 6.058 19.718 2048 493.416 41.118 46.573 36.302 41.118 10.723 30.368 4096 798.569 66.547 76.231 58.094 66.547 20.040 50.671 10624 1188.042 99.004 116.861 83.875 95.625 38.271 94.683 27554 1954.333 162.861 195.454 135.703 151.848 70.099 157.669 71468 2484.827 207.069 260.105 164.847 201.451 105.836 196.155 185364 3121.206 260.100 310.124 218.146 252.167 129.269 255.584 480774 3538.877 294.906 361.892 240.320 292.102 139.623 281.192 1246974 4098.467 341.539 435.320 267.961 336.261 162.682 320.056 3234251 4157.400 346.450 433.500 276.880 346.450 346.450 346.450 8388608 4308.890 359.074 439.721 293.218 359.074 359.074 359.074 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-6*2fix :( 15.974) 0.063 1.003 13.668 118.478 389.117 496.723 -> 170.946 -> 2051.346 MByte/s p01 ring-3*4fix :( 25.238) 0.040 0.590 8.667 69.614 306.105 412.642 -> 124.320 -> 1491.837 MByte/s p02 ring-1*12fix :( 25.027) 0.040 0.596 8.778 69.806 286.424 433.643 -> 123.794 -> 1485.530 MByte/s p03 ring-1*12fix :( 25.015) 0.040 0.595 8.805 69.479 288.752 432.984 -> 122.163 -> 1465.960 MByte/s p04 ring-1*12fix :( 25.051) 0.040 0.594 8.822 70.305 296.900 441.111 -> 125.649 -> 1507.791 MByte/s p05 ring-1*12fix :( 25.164) 0.040 0.596 8.892 69.781 304.170 425.818 -> 126.195 -> 1514.343 MByte/s p06 random-cyc-1dim :( 27.878) 0.036 0.561 8.183 57.690 253.592 351.282 -> 103.984 -> 1247.804 MByte/s p07 random-cyc-1dim :( 27.901) 0.036 0.558 8.158 57.919 240.646 335.739 -> 97.043 -> 1164.521 MByte/s p08 random-cyc-1dim :( 27.896) 0.036 0.551 8.047 57.316 211.080 280.340 -> 85.871 -> 1030.453 MByte/s p09 random-cyc-1dim :( 28.496) 0.035 0.532 7.824 55.440 145.092 174.835 -> 58.967 -> 707.598 MByte/s p10 random-cyc-1dim :( 27.784) 0.036 0.558 8.236 57.303 263.602 347.556 -> 102.150 -> 1225.797 MByte/s p11 random-cyc-1dim :( 27.918) 0.036 0.550 8.034 57.341 206.176 273.570 -> 84.502 -> 1014.021 MByte/s p12 random-cyc-1dim :( 28.297) 0.035 0.542 8.026 55.537 173.897 213.635 -> 70.171 -> 842.055 MByte/s p13 random-cyc-1dim :( 27.897) 0.036 0.557 8.213 57.576 217.819 297.020 -> 86.965 -> 1043.582 MByte/s p14 random-cyc-1dim :( 28.173) 0.035 0.555 8.136 57.133 209.596 285.521 -> 84.938 -> 1019.253 MByte/s p15 random-cyc-1dim :( 28.057) 0.036 0.548 8.033 56.725 210.654 285.128 -> 83.718 -> 1004.611 MByte/s p16 random-cyc-1dim :( 27.952) 0.036 0.550 8.078 56.917 221.978 275.737 -> 85.820 -> 1029.843 MByte/s p17 random-cyc-1dim :( 28.184) 0.035 0.544 7.949 55.803 168.521 227.512 -> 70.517 -> 846.206 MByte/s p18 random-cyc-1dim :( 28.267) 0.035 0.543 7.918 56.489 173.969 227.232 -> 70.389 -> 844.668 MByte/s p19 random-cyc-1dim :( 27.903) 0.036 0.553 8.016 56.676 218.088 294.631 -> 85.369 -> 1024.430 MByte/s p20 random-cyc-1dim :( 28.165) 0.036 0.549 8.033 56.364 212.519 290.264 -> 84.690 -> 1016.284 MByte/s p21 random-cyc-1dim :( 27.844) 0.036 0.554 8.130 57.408 212.658 282.212 -> 84.853 -> 1018.233 MByte/s p22 random-cyc-1dim :( 25.380) 0.039 0.590 8.754 67.667 268.873 366.972 -> 108.699 -> 1304.388 MByte/s p23 random-cyc-1dim :( 27.907) 0.036 0.551 8.136 57.120 212.294 272.118 -> 83.129 -> 997.546 MByte/s p24 random-cyc-1dim :( 27.828) 0.036 0.557 8.278 58.224 243.775 338.982 -> 99.812 -> 1197.743 MByte/s p25 random-cyc-1dim :( 27.823) 0.036 0.562 8.237 58.619 260.961 391.781 -> 105.132 -> 1261.588 MByte/s p26 random-cyc-1dim :( 24.892) 0.040 0.595 8.824 69.565 297.367 432.715 -> 123.207 -> 1478.490 MByte/s p27 random-cyc-1dim :( 24.985) 0.040 0.594 8.802 70.418 303.958 435.998 -> 125.450 -> 1505.406 MByte/s p28 random-cyc-1dim :( 27.912) 0.036 0.555 8.177 56.840 206.049 285.395 -> 84.757 -> 1017.079 MByte/s p29 random-cyc-1dim :( 28.017) 0.036 0.553 8.035 57.258 207.191 276.164 -> 81.609 -> 979.310 MByte/s p30 random-cyc-1dim :( 27.681) 0.036 0.564 8.255 57.496 265.074 382.230 -> 106.362 -> 1276.343 MByte/s p31 random-cyc-1dim :( 27.850) 0.036 0.556 8.167 56.905 224.197 271.256 -> 85.326 -> 1023.913 MByte/s p32 random-cyc-1dim :( 28.042) 0.036 0.555 8.110 57.170 209.281 291.651 -> 84.780 -> 1017.365 MByte/s p33 random-cyc-1dim :( 27.922) 0.036 0.556 8.097 56.553 209.344 289.502 -> 85.462 -> 1025.543 MByte/s p34 random-cyc-1dim :( 27.934) 0.036 0.552 8.042 56.090 205.515 274.609 -> 82.495 -> 989.944 MByte/s p35 random-cyc-1dim :( 28.205) 0.035 0.538 7.928 56.533 177.911 223.908 -> 71.674 -> 860.092 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 29.209) 0.034 0.523 7.655 52.373 128.623 146.454 -> 52.216 -> 626.596 MByte/s p37 best bi-section :( 11.988) 0.042 0.635 9.122 90.916 384.043 579.725 -> 161.039 -> 1932.469 MByte/s p38 worst bi-section :( 25.013) 0.020 0.299 4.520 42.642 141.923 157.031 -> 51.953 -> 623.438 MByte/s p39 one PingPong Pair :( 11.527) 0.007 0.110 1.524 17.281 58.902 100.729 -> 27.303 -> 327.640 MByte/s p40 acyclic-2dim-all :( 20.335) 0.035 0.525 7.795 62.231 265.847 342.065 -> 104.150 -> 1249.795 MByte/s p41 acyclic-3dim-all :( 20.562) 0.027 0.406 5.963 48.984 257.820 315.179 -> 96.578 -> 1158.936 MByte/s p42 cyclic-2dim-x :( 25.561) 0.039 0.584 8.703 65.764 223.613 298.921 -> 91.096 -> 1093.151 MByte/s p43 cyclic-2dim-y :( 16.101) 0.062 1.007 13.461 119.484 359.136 494.713 -> 159.492 -> 1913.902 MByte/s p44 cyclic-2dim-all :( 20.833) 0.048 0.716 10.546 83.445 308.194 374.579 -> 119.740 -> 1436.885 MByte/s p45 cyclic-3dim-x :( 28.057) 0.036 0.552 8.079 56.692 192.054 221.707 -> 73.457 -> 881.488 MByte/s p46 cyclic-3dim-y :( 28.009) 0.036 0.552 8.100 58.383 298.230 340.779 -> 106.321 -> 1275.854 MByte/s p47 cyclic-3dim-z :( 16.156) 0.062 0.983 13.473 120.100 402.237 502.189 -> 171.435 -> 2057.220 MByte/s p48 cyclic-3dim-all :( 23.848) 0.042 0.634 9.331 69.958 258.266 295.340 -> 96.761 -> 1161.136 MByte/s log_avg of all rings : 0.043 0.648 9.464 76.231 310.124 439.721 || 131.182 -> 1574.182 MByte/s log_avg of all random : 0.036 0.556 8.159 58.094 218.146 293.218 || 87.732 -> 1052.784 MByte/s log_avg(ring,random) : 0.039 0.600 8.787 66.547 260.100 359.074 || 107.279 -> 1287.352 MByte/s * size -> accumulated on all pr.: 0.473 7.205 105.442 798.569 3121.206 4308.890 || 1287.352 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 1287.352 MByte/s on 12 processes ( = 107.279 MByte/s * 12 processes) Ping-pong latency: 11.527 microsec Ping-pong bandwidth: 1208.742 MByte/s at Lmax= 8.000 MByte (MByte/s=1e6 Byte/s) (MByte=2**20 Byte) system parameters : 12 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 17:33:27 1999 Total execution wall clock time = 96 seconds SECTION-BEFF-END b_eff = 1287.352 MB/s = 107.279 * 12 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000