b_eff = 1530.664 MB/s = 95.667 * 16 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000 SECTION-HEAD-BEGIN b_eff.c, Revision 3.3 from Nov. 29, 1999 MEMORY_PER_PROCESSOR = 1024 MBytes [1M = 1024*1024] 1-dim-paterns: size = 16 2-dim-paterns: size = 4 * 4 3-dim-paterns: size = 4 * 2 * 2 Used message lengths: msglng= 1 2 4 8 16 32 64 128 256 512 1024 2048 4096 10624 27554 71468 185364 480774 1246974 3234251 8388608 Used methods: 0=Sndrcv 1=Alltoal 2=non-blk Used patterns: 0=ring-8*2fix 1=ring-4*4fix 2=ring-2*8fix 3=ring-1*16fix 4=ring-1*16fix 5=ring-1*16fix 6=random-cyc-1dim 7=random-cyc-1dim 8=random-cyc-1dim 9=random-cyc-1dim 10=random-cyc-1dim 11=random-cyc-1dim 12=random-cyc-1dim 13=random-cyc-1dim 14=random-cyc-1dim 15=random-cyc-1dim 16=random-cyc-1dim 17=random-cyc-1dim 18=random-cyc-1dim 19=random-cyc-1dim 20=random-cyc-1dim 21=random-cyc-1dim 22=random-cyc-1dim 23=random-cyc-1dim 24=random-cyc-1dim 25=random-cyc-1dim 26=random-cyc-1dim 27=random-cyc-1dim 28=random-cyc-1dim 29=random-cyc-1dim 30=random-cyc-1dim 31=random-cyc-1dim 32=random-cyc-1dim 33=random-cyc-1dim 34=random-cyc-1dim 35=random-cyc-1dim 36=worst-cyc-1dim 37=best bi-section 38=worst bi-section 39=one PingPong Pair 40=acyclic-2dim-all 41=acyclic-3dim-all 42=cyclic-2dim-x 43=cyclic-2dim-y 44=cyclic-2dim-all 45=cyclic-3dim-x 46=cyclic-3dim-y 47=cyclic-3dim-z 48=cyclic-3dim-all SECTION-HEAD-END SECTION-LOOP-BEGIN measurment loop: 122.565 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 8.6e-01 4.9e-03 4.2e-02 228 6.3e-01 3.8e-03 2.3e-02 225 6.2e-01 3.6e-03 2.3e-02 2 153 4.3e-01 2.6e-03 1.6e-02 151 4.2e-01 2.4e-03 1.6e-02 155 4.3e-01 2.5e-03 1.6e-02 4 149 4.2e-01 2.4e-03 1.6e-02 154 4.4e-01 2.5e-03 1.6e-02 153 4.3e-01 2.5e-03 1.6e-02 8 153 4.3e-01 2.5e-03 1.6e-02 151 4.2e-01 2.5e-03 1.6e-02 153 4.3e-01 2.6e-03 1.6e-02 16 152 4.6e-01 2.5e-03 1.6e-02 150 4.4e-01 2.5e-03 1.6e-02 148 4.3e-01 2.4e-03 1.6e-02 32 151 4.5e-01 2.5e-03 1.7e-02 151 4.5e-01 2.5e-03 1.7e-02 152 4.5e-01 2.6e-03 1.7e-02 64 151 4.6e-01 2.7e-03 1.7e-02 148 4.5e-01 2.6e-03 1.7e-02 148 4.5e-01 2.6e-03 1.7e-02 128 141 4.4e-01 2.6e-03 1.7e-02 142 4.5e-01 2.7e-03 1.7e-02 144 4.5e-01 2.7e-03 1.8e-02 256 134 4.2e-01 2.4e-03 1.6e-02 132 4.3e-01 2.5e-03 1.6e-02 132 4.2e-01 2.6e-03 1.6e-02 512 137 4.5e-01 2.8e-03 1.9e-02 131 4.9e-01 2.7e-03 4.2e-02 125 4.1e-01 2.5e-03 1.5e-02 1024 124 4.2e-01 2.6e-03 1.6e-02 121 4.4e-01 2.6e-03 1.8e-02 126 4.3e-01 2.8e-03 1.6e-02 2048 118 6.9e-01 3.2e-03 2.3e-02 117 8.3e-01 3.2e-03 9.1e-02 113 6.4e-01 3.2e-03 2.1e-02 4096 91 6.5e-01 3.3e-03 2.2e-02 91 7.9e-01 3.3e-03 8.9e-02 89 6.8e-01 3.2e-03 2.2e-02 10624 53 7.0e-01 3.0e-03 2.3e-02 53 7.1e-01 3.0e-03 2.4e-02 53 6.9e-01 3.0e-03 2.3e-02 27554 34 7.7e-01 3.0e-03 2.7e-02 34 7.7e-01 2.9e-03 2.8e-02 34 7.6e-01 2.9e-03 2.7e-02 71468 21 1.0e+00 3.6e-03 4.1e-02 22 1.0e+00 3.8e-03 4.3e-02 22 1.0e+00 4.0e-03 4.4e-02 185364 11 1.0e+00 4.1e-03 4.1e-02 11 9.9e-01 4.0e-03 4.0e-02 10 8.9e-01 3.6e-03 3.7e-02 480774 5 1.2e+00 4.5e-03 5.1e-02 5 1.1e+00 4.5e-03 4.9e-02 5 1.1e+00 4.4e-03 4.9e-02 1246974 2 1.1e+00 4.8e-03 8.4e-02 2 1.1e+00 5.0e-03 8.3e-02 2 1.1e+00 4.9e-03 8.6e-02 3234251 1 8.9e-01 7.2e-03 7.0e-02 M 1 7.9e-01 7.1e-03 5.6e-02 M 1 7.8e-01 7.4e-03 7.0e-02 M 8388608 1 1.8e+00 1.7e-02 1.4e-01 R 1 1.9e+00 1.8e-02 1.5e-01 R 0 0.0e+00 1.0e+30 0.0e+00 R Explanation: M = only the best method is used R = only best method and only 2 repetitions method=1 | ------- repetition 0 --------- ------- repetition 1 --------- ------- repetition 2 --------- [sec] (=Alltoal)| loop total per pattrn&mthd loop total per pattrn&mthd loop total per pattrn&mthd msg length| lng time minimum maximum lng time minimum maximum lng time minimum maximum ----------| ---- ------- ------- --------- ---- ------- ------- --------- ---- ------- ------- --------- 1 300 6.6e+00 1.2e-01 1.9e-01 27 6.0e-01 1.1e-02 2.8e-02 9 1.9e-01 3.6e-03 4.4e-03 2 150 3.3e+00 5.9e-02 8.4e-02 13 2.8e-01 5.1e-03 6.4e-03 6 1.3e-01 2.3e-03 3.1e-03 4 75 1.6e+00 3.0e-02 3.7e-02 6 1.3e-01 2.4e-03 3.0e-03 6 1.3e-01 2.3e-03 3.0e-03 8 37 8.0e-01 1.5e-02 1.8e-02 6 1.3e-01 2.4e-03 3.1e-03 6 1.3e-01 2.3e-03 3.0e-03 16 18 3.9e-01 7.2e-03 8.9e-03 6 1.3e-01 2.4e-03 3.0e-03 6 1.3e-01 2.4e-03 3.1e-03 32 9 2.0e-01 3.5e-03 4.5e-03 6 1.3e-01 2.4e-03 3.0e-03 6 1.3e-01 2.4e-03 3.0e-03 64 6 1.3e-01 2.4e-03 3.0e-03 6 1.3e-01 2.4e-03 3.0e-03 6 1.3e-01 2.4e-03 3.0e-03 128 6 1.3e-01 2.4e-03 3.0e-03 6 1.3e-01 2.4e-03 3.0e-03 6 1.3e-01 2.4e-03 3.1e-03 256 6 1.3e-01 2.4e-03 3.0e-03 6 1.3e-01 2.4e-03 3.1e-03 6 1.3e-01 2.4e-03 3.1e-03 512 6 1.4e-01 2.5e-03 4.2e-03 6 1.5e-01 2.5e-03 1.4e-02 6 1.3e-01 2.5e-03 3.3e-03 1024 6 1.3e-01 2.4e-03 3.1e-03 6 1.5e-01 2.5e-03 1.2e-02 6 1.3e-01 2.4e-03 3.1e-03 2048 6 1.6e-01 2.9e-03 4.3e-03 5 1.9e-01 2.4e-03 5.8e-02 6 1.5e-01 2.9e-03 3.7e-03 4096 5 1.4e-01 2.4e-03 3.5e-03 5 1.5e-01 2.4e-03 1.1e-02 5 1.4e-01 2.4e-03 3.6e-03 10624 4 1.5e-01 2.0e-03 4.8e-03 3 1.2e-01 1.5e-03 7.0e-03 3 1.1e-01 1.5e-03 3.0e-03 27554 3 1.6e-01 1.6e-03 5.8e-03 3 1.6e-01 1.6e-03 6.4e-03 3 1.6e-01 1.6e-03 5.2e-03 71468 3 2.8e-01 2.6e-03 1.0e-02 3 2.8e-01 2.6e-03 8.7e-03 3 2.8e-01 2.5e-03 1.0e-02 185364 2 4.3e-01 3.3e-03 2.0e-02 2 3.8e-01 3.3e-03 1.2e-02 2 3.9e-01 3.3e-03 2.2e-02 480774 1 5.1e-01 3.8e-03 2.9e-02 1 4.8e-01 3.8e-03 2.6e-02 1 4.6e-01 2.9e-03 2.2e-02 1246974 1 1.0e+00 9.1e-03 5.2e-02 1 1.0e+00 9.0e-03 4.5e-02 1 1.0e+00 9.0e-03 4.4e-02 3234251 1 2.7e-02 2.7e-02 2.7e-02 M 1 1.0e-01 5.0e-02 5.5e-02 M 1 0.0e+00 0.0e+00 0.0e+00 M 8388608 1 6.9e-02 6.9e-02 6.9e-02 R 1 2.3e-01 1.0e-01 1.3e-01 R 0 0.0e+00 1.0e+30 0.0e+00 R Explanation: M = only the best method is used R = only best method and only 2 repetitions method=2 | ------- repetition 0 --------- ------- repetition 1 --------- ------- repetition 2 --------- [sec] (=non-blk)| loop total per pattrn&mthd loop total per pattrn&mthd loop total per pattrn&mthd msg length| lng time minimum maximum lng time minimum maximum lng time minimum maximum ----------| ---- ------- ------- --------- ---- ------- ------- --------- ---- ------- ------- --------- 1 300 1.5e+00 1.1e-02 1.1e-01 96 4.5e-01 3.5e-03 1.6e-02 102 4.8e-01 3.8e-03 1.7e-02 2 150 7.3e-01 5.6e-03 2.6e-02 68 3.2e-01 2.5e-03 1.2e-02 67 3.2e-01 2.5e-03 1.1e-02 4 75 3.7e-01 2.9e-03 1.3e-02 67 3.2e-01 2.5e-03 1.2e-02 67 3.2e-01 2.5e-03 1.2e-02 8 64 3.1e-01 2.5e-03 1.1e-02 66 3.1e-01 2.4e-03 1.1e-02 67 3.2e-01 2.5e-03 1.1e-02 16 64 3.2e-01 2.5e-03 1.1e-02 68 3.3e-01 2.5e-03 1.2e-02 67 3.3e-01 2.6e-03 1.2e-02 32 63 3.2e-01 2.4e-03 1.1e-02 67 3.3e-01 2.5e-03 1.2e-02 65 3.2e-01 2.5e-03 1.2e-02 64 65 3.3e-01 2.6e-03 1.2e-02 66 3.3e-01 2.6e-03 1.2e-02 65 3.3e-01 2.5e-03 1.2e-02 128 63 3.3e-01 2.6e-03 1.2e-02 64 3.3e-01 2.6e-03 1.2e-02 64 3.3e-01 2.6e-03 1.2e-02 256 61 3.2e-01 2.4e-03 1.1e-02 61 3.3e-01 2.6e-03 1.2e-02 62 3.2e-01 2.4e-03 1.1e-02 512 62 3.5e-01 2.6e-03 2.0e-02 57 3.5e-01 2.5e-03 2.1e-02 63 3.3e-01 2.5e-03 1.2e-02 1024 58 3.2e-01 2.5e-03 1.1e-02 57 3.7e-01 2.5e-03 3.0e-02 63 3.4e-01 2.6e-03 1.2e-02 2048 59 4.3e-01 2.8e-03 1.4e-02 57 4.4e-01 2.8e-03 2.1e-02 61 4.4e-01 3.0e-03 1.4e-02 4096 53 4.6e-01 2.9e-03 1.5e-02 50 5.2e-01 2.8e-03 4.8e-02 51 4.6e-01 3.0e-03 1.5e-02 10624 34 4.6e-01 2.8e-03 1.6e-02 34 4.4e-01 2.6e-03 1.4e-02 33 4.1e-01 2.7e-03 1.3e-02 27554 23 4.9e-01 2.5e-03 1.7e-02 24 5.0e-01 2.5e-03 2.5e-02 23 4.6e-01 2.7e-03 1.7e-02 71468 17 8.2e-01 3.3e-03 3.6e-02 18 8.2e-01 3.6e-03 3.4e-02 16 7.3e-01 3.3e-03 3.1e-02 185364 9 8.6e-01 3.4e-03 3.3e-02 9 7.9e-01 3.3e-03 3.3e-02 9 7.9e-01 3.3e-03 3.6e-02 480774 5 1.1e+00 4.3e-03 5.4e-02 5 1.1e+00 4.5e-03 4.9e-02 5 1.0e+00 4.8e-03 4.8e-02 1246974 2 1.2e+00 4.7e-03 6.9e-02 2 1.1e+00 4.7e-03 6.8e-02 2 1.1e+00 4.6e-03 7.7e-02 3234251 1 4.6e-01 6.7e-03 7.3e-02 M 1 4.2e-01 6.8e-03 7.2e-02 M 1 4.6e-01 6.9e-03 4.2e-02 M 8388608 1 1.0e+00 1.8e-02 1.0e-01 R 1 9.1e-01 1.8e-02 9.5e-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 = 122.565 sec sum of max elapsed time per entries above = 122.329 sec difference to elapsed time = 0.236 sec = 0.2% sum based on fastest repetition = 105.962 sec difference to elapsed time = 16.603 sec = 13.5% CAUTION: A difference above is more than 5 %. There may be problems with the MPI implementation or processes weren't on dedicated processors SECTION-ELAPSED-END SECTION-PATTERN-BEGIN Pattern parameters: sendrecv total number messages/ messages/ unused best method calls of messages used node node&call nodes per repetition ---- ----------------- ----- ------------ --------- --------- ------ -------------- p00 ring-8*2fix 1 16 1.00 1.00 0 ( 2 2 2 ) p01 ring-4*4fix 2 32 2.00 1.00 0 ( 0 2 2 ) p02 ring-2*8fix 2 32 2.00 1.00 0 ( 0 2 0 ) p03 ring-1*16fix 2 32 2.00 1.00 0 ( 0 0 0 ) p04 ring-1*16fix 2 32 2.00 1.00 0 ( 0 0 2 ) p05 ring-1*16fix 2 32 2.00 1.00 0 ( 0 0 2 ) p06 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 0 0 ) p07 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p08 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 2 ) p09 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p10 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 0 2 ) p11 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p12 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 1 0 ) p13 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 2 ) p14 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 0 ) p15 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 0 0 ) p16 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 1 0 ) p17 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p18 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 0 0 ) p19 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 0 ) p20 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 0 0 ) p21 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 2 ) p22 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 0 ) p23 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p24 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p25 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 2 ) p26 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p27 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p28 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 2 ) p29 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 0 ) p30 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 2 ) p31 random-cyc-1dim 2 32 2.00 1.00 0 ( 2 2 2 ) p32 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p33 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 2 ) p34 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 2 2 ) p35 random-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p36 worst-cyc-1dim 2 32 2.00 1.00 0 ( 0 0 0 ) p37 best bi-section 2 16 1.00 0.50 0 ( 2 2 2 ) p38 worst bi-section 2 16 1.00 0.50 0 ( 1 2 2 ) p39 one PingPong Pair 2 2 1.00 0.50 14 ( 0 0 0 ) p40 acyclic-2dim-all 4 48 3.00 0.75 0 ( 2 2 2 ) p41 acyclic-3dim-all 6 56 3.50 0.58 0 ( 2 2 2 ) p42 cyclic-2dim-x 2 32 2.00 1.00 0 ( 0 0 0 ) p43 cyclic-2dim-y 2 32 2.00 1.00 0 ( 2 0 0 ) p44 cyclic-2dim-all 4 64 4.00 1.00 0 ( 0 2 2 ) p45 cyclic-3dim-x 2 32 2.00 1.00 0 ( 0 0 0 ) p46 cyclic-3dim-y 1 16 1.00 1.00 0 ( 2 2 2 ) p47 cyclic-3dim-z 1 16 1.00 1.00 0 ( 2 2 2 ) p48 cyclic-3dim-all 4 64 4.00 1.00 0 ( 2 0 2 ) SECTION-PATTERN-END Printing accumulated bandwidth / number of processes = bandwidth per process (in MBytes/sec = 1e6 Bytes/sec) Additionally the accumulated value processes is printed in the last column Only the cyclic and random patterns are used to compute b_eff (except p00), the other patterns are only as information. SECTION-BY-METHODS-BEGIN 2nd analysis path, only as information, last row: for all mthd: logavg_pat ( avg_L ( max_rep ( b(L) ) ) ) || logavg_pat ( max_mthd ( avg_L ( max_rep ( b(L) ) ) ) ) pattern / methods: Sndrcv Alltoal non-blk -> best method -> accumulated p00 ring-8*2fix : 160.192 69.436 141.799 -> 160.192 -> 2563.070 MByte/s p01 ring-4*4fix : 153.878 77.788 144.871 -> 153.878 -> 2462.044 MByte/s p02 ring-2*8fix : 150.248 72.705 142.541 -> 150.248 -> 2403.960 MByte/s p03 ring-1*16fix : 114.442 58.029 109.415 -> 114.442 -> 1831.070 MByte/s p04 ring-1*16fix : 108.202 51.542 97.636 -> 108.202 -> 1731.236 MByte/s p05 ring-1*16fix : 114.844 56.091 106.147 -> 114.844 -> 1837.500 MByte/s p06 random-cyc-1dim : 76.427 43.424 68.294 -> 76.427 -> 1222.826 MByte/s p07 random-cyc-1dim : 96.390 52.108 83.714 -> 96.390 -> 1542.237 MByte/s p08 random-cyc-1dim : 63.400 36.313 60.161 -> 63.400 -> 1014.402 MByte/s p09 random-cyc-1dim : 73.818 38.052 65.780 -> 73.818 -> 1181.093 MByte/s p10 random-cyc-1dim : 79.789 39.446 73.595 -> 79.789 -> 1276.618 MByte/s p11 random-cyc-1dim : 48.240 29.361 45.596 -> 48.240 -> 771.836 MByte/s p12 random-cyc-1dim : 54.644 30.352 49.703 -> 54.644 -> 874.299 MByte/s p13 random-cyc-1dim : 64.192 33.597 56.177 -> 64.192 -> 1027.074 MByte/s p14 random-cyc-1dim : 56.220 36.591 56.237 -> 56.237 -> 899.784 MByte/s p15 random-cyc-1dim : 61.567 35.912 56.984 -> 61.567 -> 985.075 MByte/s p16 random-cyc-1dim : 58.358 36.011 55.338 -> 58.358 -> 933.724 MByte/s p17 random-cyc-1dim : 50.019 29.222 48.184 -> 50.019 -> 800.303 MByte/s p18 random-cyc-1dim : 67.573 38.146 64.656 -> 67.573 -> 1081.173 MByte/s p19 random-cyc-1dim : 77.406 41.672 71.269 -> 77.406 -> 1238.493 MByte/s p20 random-cyc-1dim : 82.937 41.314 68.727 -> 82.937 -> 1326.990 MByte/s p21 random-cyc-1dim : 73.815 37.311 66.161 -> 73.815 -> 1181.038 MByte/s p22 random-cyc-1dim : 94.141 48.747 91.901 -> 94.141 -> 1506.251 MByte/s p23 random-cyc-1dim : 54.872 32.114 52.997 -> 54.872 -> 877.953 MByte/s p24 random-cyc-1dim : 85.376 44.622 78.636 -> 85.376 -> 1366.012 MByte/s p25 random-cyc-1dim : 95.880 48.732 83.736 -> 95.880 -> 1534.084 MByte/s p26 random-cyc-1dim : 47.505 27.696 45.991 -> 47.505 -> 760.081 MByte/s p27 random-cyc-1dim : 64.929 37.553 62.526 -> 64.929 -> 1038.857 MByte/s p28 random-cyc-1dim : 68.896 36.015 63.846 -> 68.896 -> 1102.343 MByte/s p29 random-cyc-1dim : 64.001 38.806 58.838 -> 64.001 -> 1024.013 MByte/s p30 random-cyc-1dim : 73.240 38.562 67.316 -> 73.240 -> 1171.840 MByte/s p31 random-cyc-1dim : 61.396 36.091 57.742 -> 61.396 -> 982.342 MByte/s p32 random-cyc-1dim : 97.067 48.116 91.359 -> 97.067 -> 1553.075 MByte/s p33 random-cyc-1dim : 61.266 37.050 57.977 -> 61.266 -> 980.258 MByte/s p34 random-cyc-1dim : 63.487 38.612 61.577 -> 63.487 -> 1015.790 MByte/s p35 random-cyc-1dim : 62.953 36.546 60.460 -> 62.953 -> 1007.247 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 40.376 27.607 36.635 -> 40.376 -> 646.009 MByte/s p37 best bi-section : 138.371 69.371 148.137 -> 148.137 -> 2370.191 MByte/s p38 worst bi-section : 29.680 25.726 37.095 -> 37.095 -> 593.515 MByte/s p39 one PingPong Pair : 18.687 5.599 5.599 -> 18.687 -> 298.985 MByte/s p40 acyclic-2dim-all : 78.609 61.146 88.151 -> 88.151 -> 1410.412 MByte/s p41 acyclic-3dim-all : 87.762 69.361 101.620 -> 101.620 -> 1625.913 MByte/s p42 cyclic-2dim-x : 70.514 40.686 64.964 -> 70.514 -> 1128.229 MByte/s p43 cyclic-2dim-y : 152.216 77.999 144.799 -> 152.216 -> 2435.450 MByte/s p44 cyclic-2dim-all : 97.586 65.445 100.898 -> 100.898 -> 1614.367 MByte/s p45 cyclic-3dim-x : 69.259 40.447 64.225 -> 69.259 -> 1108.152 MByte/s p46 cyclic-3dim-y : 159.384 67.966 143.617 -> 159.384 -> 2550.144 MByte/s p47 cyclic-3dim-z : 161.468 68.177 141.158 -> 161.468 -> 2583.487 MByte/s p48 cyclic-3dim-all : 96.069 69.317 101.503 -> 101.503 -> 1624.041 MByte/s log_avg of all rings : 131.904 63.552 122.142 || 131.904 -> 2110.468 MByte/s log_avg of all random : 67.954 37.833 63.112 || 67.955 -> 1087.274 MByte/s log_avg(ring,random) : 94.675 49.034 87.799 ||( 94.676 -> 1514.812)MByte/s * size -> accumulated on all pr.: 1514.805 784.542 1404.781 ||(1514.812)MByte/s SECTION-BY-METHODS-END SECTION-BY-REPETITIONS-BEGIN 3rd analysis path, only as information, last row: for all rep.:logavg_pat ( avg_L ( max_mthd ( b(L) ) ) ) || logavg_pat ( max_rep ( avg_L ( max_mthd ( b(L) ) ) ) ) pattern / repetition: rep.0 rep.1 rep.2 -> best repetition -> accumulated p00 ring-8*2fix : 154.666 153.209 158.964 -> 158.964 -> 2543.427 MByte/s p01 ring-4*4fix : 141.323 146.732 152.482 -> 152.482 -> 2439.705 MByte/s p02 ring-2*8fix : 146.005 147.338 148.219 -> 148.219 -> 2371.509 MByte/s p03 ring-1*16fix : 103.962 107.882 113.718 -> 113.718 -> 1819.489 MByte/s p04 ring-1*16fix : 93.328 105.181 102.291 -> 105.181 -> 1682.902 MByte/s p05 ring-1*16fix : 105.966 96.340 112.226 -> 112.226 -> 1795.618 MByte/s p06 random-cyc-1dim : 59.503 70.605 69.616 -> 70.605 -> 1129.682 MByte/s p07 random-cyc-1dim : 69.367 91.064 89.470 -> 91.064 -> 1457.016 MByte/s p08 random-cyc-1dim : 52.919 63.641 60.815 -> 63.641 -> 1018.259 MByte/s p09 random-cyc-1dim : 59.136 73.690 65.104 -> 73.690 -> 1179.040 MByte/s p10 random-cyc-1dim : 68.074 78.315 73.701 -> 78.315 -> 1253.032 MByte/s p11 random-cyc-1dim : 42.780 47.118 47.751 -> 47.751 -> 764.011 MByte/s p12 random-cyc-1dim : 48.088 41.470 54.091 -> 54.091 -> 865.450 MByte/s p13 random-cyc-1dim : 54.050 55.720 61.211 -> 61.211 -> 979.369 MByte/s p14 random-cyc-1dim : 48.234 54.236 56.345 -> 56.345 -> 901.513 MByte/s p15 random-cyc-1dim : 54.137 58.873 59.135 -> 59.135 -> 946.160 MByte/s p16 random-cyc-1dim : 53.859 46.708 56.617 -> 56.617 -> 905.878 MByte/s p17 random-cyc-1dim : 48.091 48.191 48.302 -> 48.302 -> 772.829 MByte/s p18 random-cyc-1dim : 62.156 65.457 66.084 -> 66.084 -> 1057.349 MByte/s p19 random-cyc-1dim : 65.655 71.269 73.520 -> 73.520 -> 1176.322 MByte/s p20 random-cyc-1dim : 65.270 73.305 81.579 -> 81.579 -> 1305.261 MByte/s p21 random-cyc-1dim : 70.571 70.557 66.519 -> 70.571 -> 1129.138 MByte/s p22 random-cyc-1dim : 79.620 89.571 92.846 -> 92.846 -> 1485.530 MByte/s p23 random-cyc-1dim : 52.824 53.914 53.527 -> 53.914 -> 862.623 MByte/s p24 random-cyc-1dim : 78.886 83.502 81.786 -> 83.502 -> 1336.027 MByte/s p25 random-cyc-1dim : 66.258 91.929 84.438 -> 91.929 -> 1470.860 MByte/s p26 random-cyc-1dim : 44.582 46.767 47.073 -> 47.073 -> 753.167 MByte/s p27 random-cyc-1dim : 60.743 62.042 64.776 -> 64.776 -> 1036.421 MByte/s p28 random-cyc-1dim : 58.404 68.235 65.078 -> 68.235 -> 1091.758 MByte/s p29 random-cyc-1dim : 57.216 59.704 60.133 -> 60.133 -> 962.124 MByte/s p30 random-cyc-1dim : 71.611 66.341 66.673 -> 71.611 -> 1145.768 MByte/s p31 random-cyc-1dim : 58.550 60.383 60.491 -> 60.491 -> 967.854 MByte/s p32 random-cyc-1dim : 86.374 96.366 94.130 -> 96.366 -> 1541.848 MByte/s p33 random-cyc-1dim : 53.800 59.038 60.763 -> 60.763 -> 972.208 MByte/s p34 random-cyc-1dim : 61.577 57.315 63.301 -> 63.301 -> 1012.809 MByte/s p35 random-cyc-1dim : 62.571 60.311 59.486 -> 62.571 -> 1001.137 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 38.956 39.839 40.092 -> 40.092 -> 641.467 MByte/s p37 best bi-section : 145.534 141.972 144.137 -> 145.534 -> 2328.539 MByte/s p38 worst bi-section : 35.891 35.359 35.632 -> 35.891 -> 574.254 MByte/s p39 one PingPong Pair : 18.537 18.152 17.460 -> 18.537 -> 296.590 MByte/s p40 acyclic-2dim-all : 82.565 85.721 84.462 -> 85.721 -> 1371.543 MByte/s p41 acyclic-3dim-all : 98.277 101.539 100.687 -> 101.539 -> 1624.622 MByte/s p42 cyclic-2dim-x : 65.855 69.513 69.292 -> 69.513 -> 1112.216 MByte/s p43 cyclic-2dim-y : 154.105 152.169 151.610 -> 154.105 -> 2465.673 MByte/s p44 cyclic-2dim-all : 96.731 101.018 100.510 -> 101.018 -> 1616.281 MByte/s p45 cyclic-3dim-x : 67.470 68.079 66.740 -> 68.079 -> 1089.263 MByte/s p46 cyclic-3dim-y : 160.640 157.274 152.638 -> 160.640 -> 2570.239 MByte/s p47 cyclic-3dim-z : 161.617 161.281 157.576 -> 161.617 -> 2585.873 MByte/s p48 cyclic-3dim-all : 101.845 97.942 102.978 -> 102.978 -> 1647.645 MByte/s log_avg of all rings : 121.900 123.918 129.379 || 129.981 -> 2079.700 MByte/s log_avg of all random : 59.651 64.009 65.014 || 66.386 -> 1062.180 MByte/s log_avg(ring,random) : 85.273 89.061 91.714 ||( 92.892 -> 1486.275)MByte/s * size -> accumulated on all pr.: 1364.372 1424.981 1467.420 ||(1486.275)MByte/s SECTION-BY-REPETITIONS-END SECTION-BY-MTHD-MSGLNG-BEGIN 4th analysis path, only as information, last 3 rows: for all methods and some msglng L: logavg_pat ( max_rep ( b(L) ) ) for all methods: logavg_pat ( avg_L ( max_rep ( b(L) ) ) ) pattern & method / msg: 1 16 256 4096 185364 8388608 -> average -> accumulated (latency of one sendrecv (or equivalent method) in microsec) p00 ring-8*2fix p00 method 0 =Sndrcv :( 16.840) 0.059 0.935 12.553 112.411 406.501 450.853 -> 160.192 -> 2563.070 MByte/s p00 method 1 =Alltoal :(477.186) 0.002 0.034 0.543 8.445 111.462 450.853 -> 69.436 -> 1110.972 MByte/s p00 method 2 =non-blk :( 38.636) 0.026 0.413 6.053 72.112 356.013 450.853 -> 141.799 -> 2268.785 MByte/s p01 ring-4*4fix p01 method 0 =Sndrcv :( 16.757) 0.060 0.932 12.781 112.898 364.303 454.298 -> 153.878 -> 2462.044 MByte/s p01 method 1 =Alltoal :(233.054) 0.004 0.067 1.087 16.450 147.640 454.298 -> 77.788 -> 1244.613 MByte/s p01 method 2 =non-blk :( 37.036) 0.027 0.434 6.396 76.171 363.536 454.298 -> 144.871 -> 2317.935 MByte/s p02 ring-2*8fix p02 method 0 =Sndrcv :( 17.454) 0.057 0.927 12.486 111.839 399.966 457.069 -> 150.248 -> 2403.960 MByte/s p02 method 1 =Alltoal :(233.114) 0.004 0.068 1.073 16.331 103.527 457.069 -> 72.705 -> 1163.283 MByte/s p02 method 2 =non-blk :( 36.370) 0.027 0.432 6.595 76.588 380.275 457.069 -> 142.541 -> 2280.657 MByte/s p03 ring-1*16fix p03 method 0 =Sndrcv :( 25.377) 0.039 0.584 8.622 64.999 275.913 372.901 -> 114.442 -> 1831.070 MByte/s p03 method 1 =Alltoal :(235.370) 0.004 0.069 1.086 16.450 89.983 372.901 -> 58.029 -> 928.466 MByte/s p03 method 2 =non-blk :( 46.219) 0.022 0.332 5.122 51.339 262.163 372.901 -> 109.415 -> 1750.639 MByte/s p04 ring-1*16fix p04 method 0 =Sndrcv :( 25.285) 0.040 0.589 8.642 65.358 278.762 382.038 -> 108.202 -> 1731.236 MByte/s p04 method 1 =Alltoal :(233.001) 0.004 0.067 1.083 16.345 90.147 382.038 -> 51.542 -> 824.672 MByte/s p04 method 2 =non-blk :( 45.583) 0.022 0.338 5.038 50.049 192.988 382.038 -> 97.636 -> 1562.169 MByte/s p05 ring-1*16fix p05 method 0 =Sndrcv :( 25.430) 0.039 0.590 8.642 65.784 306.384 361.719 -> 114.844 -> 1837.500 MByte/s p05 method 1 =Alltoal :(233.498) 0.004 0.068 1.086 16.463 67.670 361.719 -> 56.091 -> 897.452 MByte/s p05 method 2 =non-blk :( 45.599) 0.022 0.342 5.013 50.656 247.866 361.719 -> 106.147 -> 1698.347 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 33.916) 0.029 0.460 6.807 43.705 199.605 242.162 -> 76.427 -> 1222.826 MByte/s p06 method 1 =Alltoal :(221.445) 0.005 0.071 1.130 14.013 82.412 242.162 -> 43.424 -> 694.778 MByte/s p06 method 2 =non-blk :( 56.000) 0.018 0.275 4.120 37.614 141.104 242.162 -> 68.294 -> 1092.706 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 28.772) 0.035 0.523 7.782 54.169 255.710 309.800 -> 96.390 -> 1542.237 MByte/s p07 method 1 =Alltoal :(245.333) 0.004 0.065 1.042 15.126 92.763 309.800 -> 52.108 -> 833.729 MByte/s p07 method 2 =non-blk :( 49.943) 0.020 0.309 4.670 44.128 245.425 309.800 -> 83.714 -> 1339.427 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 29.329) 0.034 0.505 7.526 49.477 153.383 182.017 -> 63.400 -> 1014.402 MByte/s p08 method 1 =Alltoal :(213.279) 0.005 0.074 1.184 13.745 77.412 182.017 -> 36.313 -> 581.010 MByte/s p08 method 2 =non-blk :( 50.823) 0.020 0.304 4.549 42.899 168.097 182.017 -> 60.161 -> 962.570 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 28.904) 0.035 0.524 7.741 52.683 189.215 221.833 -> 73.818 -> 1181.093 MByte/s p09 method 1 =Alltoal :(220.482) 0.005 0.072 1.156 15.053 76.008 221.833 -> 38.052 -> 608.836 MByte/s p09 method 2 =non-blk :( 49.245) 0.020 0.315 4.623 42.100 197.207 221.833 -> 65.780 -> 1052.488 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 29.060) 0.034 0.512 7.681 53.573 207.249 214.362 -> 79.789 -> 1276.618 MByte/s p10 method 1 =Alltoal :(217.870) 0.005 0.072 1.156 14.540 74.264 214.362 -> 39.446 -> 631.130 MByte/s p10 method 2 =non-blk :( 49.917) 0.020 0.309 4.668 42.648 207.948 214.362 -> 73.595 -> 1177.524 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 29.723) 0.034 0.501 7.619 49.435 112.206 138.757 -> 48.240 -> 771.836 MByte/s p11 method 1 =Alltoal :(198.556) 0.005 0.079 1.262 12.615 68.947 138.757 -> 29.361 -> 469.768 MByte/s p11 method 2 =non-blk :( 51.234) 0.020 0.299 4.480 40.157 120.649 138.757 -> 45.596 -> 729.532 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 29.175) 0.034 0.513 7.686 51.834 133.288 168.656 -> 54.644 -> 874.299 MByte/s p12 method 1 =Alltoal :(210.557) 0.005 0.076 1.199 12.966 61.282 168.656 -> 30.352 -> 485.637 MByte/s p12 method 2 =non-blk :( 50.437) 0.020 0.303 4.527 37.960 126.942 168.656 -> 49.703 -> 795.251 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 28.449) 0.035 0.542 7.979 53.401 160.925 185.735 -> 64.192 -> 1027.074 MByte/s p13 method 1 =Alltoal :(223.833) 0.004 0.071 1.122 14.192 73.521 185.735 -> 33.597 -> 537.556 MByte/s p13 method 2 =non-blk :( 49.193) 0.020 0.316 4.710 42.860 161.123 185.735 -> 56.177 -> 898.839 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 34.114) 0.029 0.452 6.702 43.394 141.888 147.467 -> 56.220 -> 899.526 MByte/s p14 method 1 =Alltoal :(214.835) 0.005 0.074 1.155 12.824 91.695 147.467 -> 36.591 -> 585.451 MByte/s p14 method 2 =non-blk :( 55.234) 0.018 0.277 4.192 36.948 139.278 147.467 -> 56.237 -> 899.784 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 29.314) 0.034 0.509 7.704 46.477 158.369 190.345 -> 61.567 -> 985.075 MByte/s p15 method 1 =Alltoal :(213.200) 0.005 0.075 1.169 14.188 69.745 190.345 -> 35.912 -> 574.600 MByte/s p15 method 2 =non-blk :( 51.667) 0.019 0.295 4.498 40.775 140.344 190.345 -> 56.984 -> 911.739 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 34.163) 0.029 0.456 6.744 43.407 163.788 151.910 -> 58.358 -> 933.724 MByte/s p16 method 1 =Alltoal :(213.497) 0.005 0.074 1.161 13.307 88.701 151.910 -> 36.011 -> 576.178 MByte/s p16 method 2 =non-blk :( 54.761) 0.018 0.275 4.166 37.176 142.563 151.910 -> 55.338 -> 885.406 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 29.940) 0.033 0.501 7.488 49.418 120.358 156.400 -> 50.019 -> 800.303 MByte/s p17 method 1 =Alltoal :(202.219) 0.005 0.079 1.244 12.619 66.930 156.400 -> 29.222 -> 467.553 MByte/s p17 method 2 =non-blk :( 51.557) 0.019 0.294 4.448 40.665 129.963 156.400 -> 48.184 -> 770.949 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 29.083) 0.034 0.513 7.558 49.688 169.184 209.409 -> 67.573 -> 1081.173 MByte/s p18 method 1 =Alltoal :(216.464) 0.005 0.075 1.177 13.970 84.094 209.409 -> 38.146 -> 610.336 MByte/s p18 method 2 =non-blk :( 50.323) 0.020 0.302 4.542 43.408 162.632 209.409 -> 64.656 -> 1034.502 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 28.742) 0.035 0.529 7.862 53.590 198.424 252.821 -> 77.406 -> 1238.493 MByte/s p19 method 1 =Alltoal :(234.924) 0.004 0.068 1.082 14.733 83.648 252.821 -> 41.672 -> 666.756 MByte/s p19 method 2 =non-blk :( 48.734) 0.021 0.308 4.729 42.082 195.669 252.821 -> 71.269 -> 1140.310 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 33.549) 0.030 0.463 6.887 44.106 256.140 209.524 -> 82.937 -> 1326.990 MByte/s p20 method 1 =Alltoal :(231.871) 0.004 0.069 1.092 14.163 99.085 209.524 -> 41.314 -> 661.028 MByte/s p20 method 2 =non-blk :( 56.547) 0.018 0.271 4.093 34.944 160.697 209.524 -> 68.727 -> 1099.635 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 28.186) 0.035 0.547 8.034 53.697 188.044 193.945 -> 73.815 -> 1181.038 MByte/s p21 method 1 =Alltoal :(218.703) 0.005 0.074 1.163 14.755 77.035 193.945 -> 37.311 -> 596.975 MByte/s p21 method 2 =non-blk :( 48.672) 0.021 0.316 4.693 43.062 187.869 193.945 -> 66.161 -> 1058.575 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 28.587) 0.035 0.525 7.908 56.625 233.684 283.069 -> 94.141 -> 1506.251 MByte/s p22 method 1 =Alltoal :(225.504) 0.004 0.071 1.114 15.120 89.042 283.069 -> 48.747 -> 779.944 MByte/s p22 method 2 =non-blk :( 49.026) 0.020 0.313 4.723 42.815 247.923 283.069 -> 91.901 -> 1470.422 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 34.231) 0.029 0.455 6.718 44.126 137.743 171.248 -> 54.872 -> 877.953 MByte/s p23 method 1 =Alltoal :(206.610) 0.005 0.077 1.196 12.473 71.583 171.248 -> 32.114 -> 513.818 MByte/s p23 method 2 =non-blk :( 55.468) 0.018 0.277 4.123 37.452 140.901 171.248 -> 52.997 -> 847.952 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 28.928) 0.035 0.515 7.630 55.703 230.767 278.077 -> 85.376 -> 1366.012 MByte/s p24 method 1 =Alltoal :(209.610) 0.005 0.075 1.187 14.771 84.864 278.077 -> 44.622 -> 713.956 MByte/s p24 method 2 =non-blk :( 47.526) 0.021 0.327 4.846 46.882 225.094 278.077 -> 78.636 -> 1258.174 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 28.530) 0.035 0.520 7.819 54.244 278.784 313.517 -> 95.880 -> 1534.084 MByte/s p25 method 1 =Alltoal :(238.948) 0.004 0.067 1.060 14.998 78.669 313.517 -> 48.732 -> 779.710 MByte/s p25 method 2 =non-blk :( 48.891) 0.020 0.315 4.779 42.256 215.456 313.517 -> 83.736 -> 1339.782 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 34.629) 0.029 0.451 6.720 40.767 116.372 151.007 -> 47.505 -> 760.081 MByte/s p26 method 1 =Alltoal :(199.113) 0.005 0.080 1.251 12.012 62.396 151.007 -> 27.696 -> 443.143 MByte/s p26 method 2 =non-blk :( 55.948) 0.018 0.271 4.135 36.157 124.638 151.007 -> 45.991 -> 735.856 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 34.026) 0.029 0.458 6.800 45.788 163.706 215.457 -> 64.929 -> 1038.857 MByte/s p27 method 1 =Alltoal :(214.775) 0.005 0.074 1.162 13.713 82.946 215.457 -> 37.553 -> 600.849 MByte/s p27 method 2 =non-blk :( 55.334) 0.018 0.274 4.188 38.146 170.834 215.457 -> 62.526 -> 1000.419 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 28.801) 0.035 0.522 7.656 54.114 181.640 196.209 -> 68.896 -> 1102.343 MByte/s p28 method 1 =Alltoal :(227.557) 0.004 0.070 1.115 14.808 81.024 196.209 -> 36.015 -> 576.242 MByte/s p28 method 2 =non-blk :( 50.093) 0.020 0.309 4.600 44.204 187.069 196.209 -> 63.846 -> 1021.539 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 29.037) 0.034 0.523 7.702 53.712 175.572 172.127 -> 64.001 -> 1024.013 MByte/s p29 method 1 =Alltoal :(220.895) 0.005 0.072 1.137 14.702 89.257 172.127 -> 38.806 -> 620.901 MByte/s p29 method 2 =non-blk :( 50.989) 0.020 0.302 4.555 41.971 143.779 172.127 -> 58.838 -> 941.414 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 28.513) 0.035 0.520 7.775 54.004 161.717 205.691 -> 73.240 -> 1171.840 MByte/s p30 method 1 =Alltoal :(241.762) 0.004 0.067 1.052 14.004 78.286 205.691 -> 38.562 -> 616.994 MByte/s p30 method 2 =non-blk :( 49.797) 0.020 0.309 4.656 43.129 138.452 205.691 -> 67.316 -> 1077.051 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 34.022) 0.029 0.457 6.781 45.386 159.741 191.718 -> 61.396 -> 982.342 MByte/s p31 method 1 =Alltoal :(211.166) 0.005 0.075 1.176 13.251 71.014 191.718 -> 36.091 -> 577.459 MByte/s p31 method 2 =non-blk :( 55.348) 0.018 0.276 4.159 36.989 142.344 191.718 -> 57.742 -> 923.865 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 28.739) 0.035 0.523 7.722 57.287 239.447 321.976 -> 97.067 -> 1553.075 MByte/s p32 method 1 =Alltoal :(225.908) 0.004 0.071 1.116 14.639 72.677 321.976 -> 48.116 -> 769.856 MByte/s p32 method 2 =non-blk :( 48.969) 0.020 0.313 4.705 46.293 238.121 321.976 -> 91.359 -> 1461.741 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 34.122) 0.029 0.457 6.767 45.161 149.680 191.819 -> 61.266 -> 980.258 MByte/s p33 method 1 =Alltoal :(210.053) 0.005 0.076 1.199 13.478 85.412 191.819 -> 37.050 -> 592.792 MByte/s p33 method 2 =non-blk :( 55.120) 0.018 0.275 4.140 36.510 167.103 191.819 -> 57.977 -> 927.635 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 29.333) 0.034 0.525 7.719 53.187 154.986 206.537 -> 63.487 -> 1015.790 MByte/s p34 method 1 =Alltoal :(213.235) 0.005 0.076 1.185 13.984 76.047 206.537 -> 38.612 -> 617.784 MByte/s p34 method 2 =non-blk :( 50.391) 0.020 0.305 4.564 42.323 155.521 206.537 -> 61.577 -> 985.226 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 28.469) 0.035 0.540 7.893 54.442 163.460 206.667 -> 62.953 -> 1007.247 MByte/s p35 method 1 =Alltoal :(221.259) 0.005 0.073 1.148 14.188 78.653 206.667 -> 36.546 -> 584.728 MByte/s p35 method 2 =non-blk :( 50.104) 0.020 0.311 4.544 43.275 157.326 206.667 -> 60.460 -> 967.358 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 34.624) 0.029 0.444 6.645 42.337 101.194 116.121 -> 40.376 -> 646.009 MByte/s p36 method 1 =Alltoal :(197.148) 0.005 0.080 1.276 11.686 70.307 116.121 -> 27.607 -> 441.706 MByte/s p36 method 2 =non-blk :( 56.287) 0.018 0.273 4.093 35.990 96.939 116.121 -> 36.635 -> 586.159 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 12.327) 0.041 0.631 8.954 83.536 327.551 470.452 -> 138.371 -> 2213.941 MByte/s p37 method 1 =Alltoal :(236.523) 0.002 0.034 0.541 8.456 106.074 470.452 -> 69.371 -> 1109.944 MByte/s p37 method 2 =non-blk :( 18.343) 0.027 0.430 6.441 71.482 370.402 470.452 -> 148.137 -> 2370.191 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 28.518) 0.018 0.260 3.854 29.893 79.795 110.812 -> 29.680 -> 474.886 MByte/s p38 method 1 =Alltoal :(233.889) 0.002 0.034 0.537 7.541 41.664 110.812 -> 25.726 -> 411.611 MByte/s p38 method 2 =non-blk :( 31.589) 0.016 0.244 3.821 31.903 96.953 110.812 -> 37.095 -> 593.515 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 11.811) 0.005 0.081 1.150 12.425 48.099 60.611 -> 18.687 -> 298.985 MByte/s p39 method 1 =Alltoal :(-------) 0.000 0.000 0.000 0.000 0.000 60.611 -> 5.599 -> 89.582 MByte/s p39 method 2 =non-blk :(-------) 0.000 0.000 0.000 0.000 0.000 60.611 -> 5.599 -> 89.582 MByte/s p40 acyclic-2dim-all p40 method 0 =Sndrcv :( 21.997) 0.034 0.515 7.589 58.428 214.558 246.369 -> 78.609 -> 1257.745 MByte/s p40 method 1 =Alltoal :(119.243) 0.006 0.102 1.591 18.980 133.195 246.369 -> 61.146 -> 978.343 MByte/s p40 method 2 =non-blk :( 38.816) 0.019 0.300 4.513 46.173 251.854 246.369 -> 88.151 -> 1410.412 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 16.997) 0.034 0.520 7.551 59.796 222.308 295.127 -> 87.762 -> 1404.190 MByte/s p41 method 1 =Alltoal :( 78.815) 0.007 0.117 1.841 21.722 170.862 295.127 -> 69.361 -> 1109.773 MByte/s p41 method 2 =non-blk :( 26.317) 0.022 0.344 5.188 51.422 275.892 295.127 -> 101.620 -> 1625.913 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 29.316) 0.034 0.478 7.104 50.410 178.211 222.947 -> 70.514 -> 1128.229 MByte/s p42 method 1 =Alltoal :(235.798) 0.004 0.068 1.066 12.983 84.486 222.947 -> 40.686 -> 650.976 MByte/s p42 method 2 =non-blk :( 53.427) 0.019 0.295 4.366 41.640 180.082 222.947 -> 64.964 -> 1039.427 MByte/s p43 cyclic-2dim-y p43 method 0 =Sndrcv :( 16.662) 0.060 0.959 13.234 112.495 379.417 469.385 -> 152.216 -> 2435.450 MByte/s p43 method 1 =Alltoal :(233.557) 0.004 0.068 1.081 16.549 122.251 469.385 -> 77.999 -> 1247.989 MByte/s p43 method 2 =non-blk :( 36.677) 0.027 0.429 6.448 76.017 376.161 469.385 -> 144.799 -> 2316.788 MByte/s p44 cyclic-2dim-all p44 method 0 =Sndrcv :( 22.620) 0.044 0.659 9.707 68.401 246.495 338.506 -> 97.586 -> 1561.372 MByte/s p44 method 1 =Alltoal :(118.802) 0.008 0.136 2.111 24.975 148.514 338.506 -> 65.445 -> 1047.113 MByte/s p44 method 2 =non-blk :( 41.950) 0.024 0.368 5.613 58.673 280.513 338.506 -> 100.898 -> 1614.367 MByte/s p45 cyclic-3dim-x p45 method 0 =Sndrcv :( 30.395) 0.033 0.479 7.290 49.467 178.235 216.171 -> 69.259 -> 1108.152 MByte/s p45 method 1 =Alltoal :(233.776) 0.004 0.067 1.040 12.913 77.195 216.171 -> 40.447 -> 647.149 MByte/s p45 method 2 =non-blk :( 53.480) 0.019 0.292 4.453 44.254 185.663 216.171 -> 64.225 -> 1027.601 MByte/s p46 cyclic-3dim-y p46 method 0 =Sndrcv :( 16.887) 0.059 0.939 12.524 111.869 399.179 464.612 -> 159.384 -> 2550.144 MByte/s p46 method 1 =Alltoal :(466.877) 0.002 0.034 0.544 8.380 100.116 464.612 -> 67.966 -> 1087.457 MByte/s p46 method 2 =non-blk :( 38.113) 0.026 0.417 6.237 71.316 359.617 464.612 -> 143.617 -> 2297.872 MByte/s p47 cyclic-3dim-z p47 method 0 =Sndrcv :( 17.127) 0.058 0.925 12.834 111.765 408.952 466.163 -> 161.468 -> 2583.487 MByte/s p47 method 1 =Alltoal :(467.234) 0.002 0.034 0.544 8.421 109.230 466.163 -> 68.177 -> 1090.829 MByte/s p47 method 2 =non-blk :( 38.333) 0.026 0.414 6.207 71.342 337.917 466.163 -> 141.158 -> 2258.527 MByte/s p48 cyclic-3dim-all p48 method 0 =Sndrcv :( 22.764) 0.044 0.657 9.466 68.707 243.204 327.725 -> 96.069 -> 1537.099 MByte/s p48 method 1 =Alltoal :(116.974) 0.009 0.135 2.117 24.315 172.210 327.725 -> 69.317 -> 1109.078 MByte/s p48 method 2 =non-blk :( 42.257) 0.024 0.363 5.546 58.817 304.402 327.725 -> 101.503 -> 1624.041 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.048 0.740 10.433 85.717 334.300 411.060 || 131.904 -> 2110.468 MByte/s - ring, method 1 = Alltoal: 0.004 0.060 0.965 14.688 98.889 411.060 || 63.552 -> 1016.827 MByte/s - ring, method 2 = non-blk: 0.024 0.379 5.664 61.622 291.791 411.060 || 122.142 -> 1954.269 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.033 0.500 7.432 49.995 173.695 204.035 || 67.954 -> 1087.264 MByte/s - random, method 1 = Alltoal: 0.005 0.073 1.152 13.938 78.477 204.035 || 37.833 -> 605.321 MByte/s - random, method 2 = non-blk: 0.019 0.298 4.475 40.807 165.724 204.035 || 63.112 -> 1009.794 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.040 0.608 8.806 65.463 240.969 289.604 || 94.675 -> 1514.805 MByte/s - average, method 1 = Alltoal: 0.004 0.066 1.054 14.308 88.094 289.604 || 49.034 -> 784.542 MByte/s - average, method 2 = non-blk: 0.022 0.336 5.034 50.146 219.902 289.604 || 87.799 -> 1404.781 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 0.637 9.735 140.897 1047.408 3855.509 4633.666 || 1514.805 MByte/s - accumulated, mthd 1 = Alltoal: 0.067 1.062 16.870 228.930 1409.501 4633.666 || 784.542 MByte/s - accumulated, mthd 2 = non-blk: 0.347 5.375 80.551 802.336 3518.427 4633.666 || 1404.781 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 0.637 0.040 0.048 0.033 0.040 0.004 0.022 2 1.272 0.080 0.097 0.065 0.080 0.008 0.043 4 2.525 0.158 0.193 0.129 0.158 0.017 0.085 8 5.097 0.319 0.387 0.262 0.319 0.033 0.173 16 9.735 0.608 0.740 0.500 0.608 0.066 0.336 32 19.238 1.202 1.465 0.987 1.202 0.133 0.665 64 37.520 2.345 2.834 1.940 2.345 0.266 1.316 128 71.389 4.462 5.317 3.744 4.462 0.529 2.564 256 140.897 8.806 10.433 7.432 8.806 1.054 5.034 512 275.291 17.206 20.388 14.520 17.206 2.090 9.906 1024 524.193 32.762 38.135 28.146 32.762 4.150 19.399 2048 667.894 41.743 54.018 32.258 41.743 7.619 30.153 4096 1047.408 65.463 85.717 49.995 65.463 14.308 50.146 10624 1588.786 99.299 132.480 74.429 97.763 28.826 90.547 27554 2530.981 158.186 216.494 115.583 152.457 54.097 147.930 71468 3123.332 195.208 282.051 135.104 194.469 72.483 179.584 185364 3895.748 243.484 334.300 177.339 240.969 88.094 219.902 480774 4121.394 257.587 364.595 181.986 249.264 77.485 240.821 1246974 4688.525 293.033 426.570 201.299 290.892 87.062 254.679 3234251 4599.679 287.480 395.458 208.985 287.480 287.480 287.480 8388608 4633.666 289.604 411.060 204.035 289.604 289.604 289.604 SECTION-BY-MSGLNG-END SECTION-BY-PATTERN-MSGLNG-BEGIN official (1st) analysis path -- see last column: (only columns of some message lengths are printed) logavg_pat ( avg_L ( max_mthd ( max_rep ( b(L) ) ) ) ) pattern / msg-length: 1 16 256 4096 185364 8388608 -> average -> accumulated (latency of one sendrecv (or equivalent method) in microsec) p00 ring-8*2fix :( 16.840) 0.059 0.935 12.553 112.411 406.501 450.853 -> 161.297 -> 2580.750 MByte/s p01 ring-4*4fix :( 16.757) 0.060 0.932 12.781 112.898 364.303 454.298 -> 154.718 -> 2475.496 MByte/s p02 ring-2*8fix :( 17.454) 0.057 0.927 12.486 111.839 399.966 457.069 -> 153.218 -> 2451.482 MByte/s p03 ring-1*16fix :( 25.377) 0.039 0.584 8.622 64.999 275.913 372.901 -> 116.569 -> 1865.107 MByte/s p04 ring-1*16fix :( 25.285) 0.040 0.589 8.642 65.358 278.762 382.038 -> 108.202 -> 1731.236 MByte/s p05 ring-1*16fix :( 25.430) 0.039 0.590 8.642 65.784 306.384 361.719 -> 114.844 -> 1837.500 MByte/s p06 random-cyc-1dim :( 33.916) 0.029 0.460 6.807 43.705 199.605 242.162 -> 76.427 -> 1222.826 MByte/s p07 random-cyc-1dim :( 28.772) 0.035 0.523 7.782 54.169 255.710 309.800 -> 96.478 -> 1543.653 MByte/s p08 random-cyc-1dim :( 29.329) 0.034 0.505 7.526 49.477 168.097 182.017 -> 64.960 -> 1039.360 MByte/s p09 random-cyc-1dim :( 28.904) 0.035 0.524 7.741 52.683 197.207 221.833 -> 74.366 -> 1189.849 MByte/s p10 random-cyc-1dim :( 29.060) 0.034 0.512 7.681 53.573 207.948 214.362 -> 80.144 -> 1282.308 MByte/s p11 random-cyc-1dim :( 29.723) 0.034 0.501 7.619 49.435 120.649 138.757 -> 49.656 -> 794.490 MByte/s p12 random-cyc-1dim :( 29.175) 0.034 0.513 7.686 51.834 133.288 168.656 -> 55.027 -> 880.438 MByte/s p13 random-cyc-1dim :( 28.449) 0.035 0.542 7.979 53.401 161.123 185.735 -> 64.356 -> 1029.691 MByte/s p14 random-cyc-1dim :( 34.114) 0.029 0.452 6.702 43.394 141.888 147.467 -> 58.124 -> 929.990 MByte/s p15 random-cyc-1dim :( 29.314) 0.034 0.509 7.704 46.477 158.369 190.345 -> 62.154 -> 994.465 MByte/s p16 random-cyc-1dim :( 34.163) 0.029 0.456 6.744 43.407 163.788 151.910 -> 58.658 -> 938.523 MByte/s p17 random-cyc-1dim :( 29.940) 0.033 0.501 7.488 49.418 129.963 156.400 -> 51.570 -> 825.121 MByte/s p18 random-cyc-1dim :( 29.083) 0.034 0.513 7.558 49.688 169.184 209.409 -> 68.438 -> 1095.012 MByte/s p19 random-cyc-1dim :( 28.742) 0.035 0.529 7.862 53.590 198.424 252.821 -> 77.736 -> 1243.771 MByte/s p20 random-cyc-1dim :( 33.549) 0.030 0.463 6.887 44.106 256.140 209.524 -> 82.937 -> 1326.990 MByte/s p21 random-cyc-1dim :( 28.186) 0.035 0.547 8.034 53.697 188.044 193.945 -> 73.897 -> 1182.348 MByte/s p22 random-cyc-1dim :( 28.587) 0.035 0.525 7.908 56.625 247.923 283.069 -> 95.743 -> 1531.888 MByte/s p23 random-cyc-1dim :( 34.231) 0.029 0.455 6.718 44.126 140.901 171.248 -> 55.920 -> 894.720 MByte/s p24 random-cyc-1dim :( 28.928) 0.035 0.515 7.630 55.703 230.767 278.077 -> 85.768 -> 1372.290 MByte/s p25 random-cyc-1dim :( 28.530) 0.035 0.520 7.819 54.244 278.784 313.517 -> 96.503 -> 1544.051 MByte/s p26 random-cyc-1dim :( 34.629) 0.029 0.451 6.720 40.767 124.638 151.007 -> 48.948 -> 783.170 MByte/s p27 random-cyc-1dim :( 34.026) 0.029 0.458 6.800 45.788 170.834 215.457 -> 66.600 -> 1065.596 MByte/s p28 random-cyc-1dim :( 28.801) 0.035 0.522 7.656 54.114 187.069 196.209 -> 69.424 -> 1110.784 MByte/s p29 random-cyc-1dim :( 29.037) 0.034 0.523 7.702 53.712 175.572 172.127 -> 64.459 -> 1031.344 MByte/s p30 random-cyc-1dim :( 28.513) 0.035 0.520 7.775 54.004 161.717 205.691 -> 74.437 -> 1190.989 MByte/s p31 random-cyc-1dim :( 34.022) 0.029 0.457 6.781 45.386 159.741 191.718 -> 61.806 -> 988.897 MByte/s p32 random-cyc-1dim :( 28.739) 0.035 0.523 7.722 57.287 239.447 321.976 -> 97.822 -> 1565.154 MByte/s p33 random-cyc-1dim :( 34.122) 0.029 0.457 6.767 45.161 167.103 191.819 -> 62.806 -> 1004.892 MByte/s p34 random-cyc-1dim :( 29.333) 0.034 0.525 7.719 53.187 155.521 206.537 -> 64.704 -> 1035.263 MByte/s p35 random-cyc-1dim :( 28.469) 0.035 0.540 7.893 54.442 163.460 206.667 -> 63.578 -> 1017.241 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 34.624) 0.029 0.444 6.645 42.337 101.194 116.121 -> 40.890 -> 654.238 MByte/s p37 best bi-section :( 12.327) 0.041 0.631 8.954 83.536 370.402 470.452 -> 149.836 -> 2397.379 MByte/s p38 worst bi-section :( 28.518) 0.018 0.260 3.854 31.903 96.953 110.812 -> 37.359 -> 597.739 MByte/s p39 one PingPong Pair :( 11.811) 0.005 0.081 1.150 12.425 48.099 60.611 -> 18.687 -> 298.985 MByte/s p40 acyclic-2dim-all :( 21.997) 0.034 0.515 7.589 58.428 251.854 246.369 -> 90.255 -> 1444.082 MByte/s p41 acyclic-3dim-all :( 16.997) 0.034 0.520 7.551 59.796 275.892 295.127 -> 103.150 -> 1650.399 MByte/s p42 cyclic-2dim-x :( 29.316) 0.034 0.478 7.104 50.410 180.082 222.947 -> 71.521 -> 1144.333 MByte/s p43 cyclic-2dim-y :( 16.662) 0.060 0.959 13.234 112.495 379.417 469.385 -> 155.578 -> 2489.240 MByte/s p44 cyclic-2dim-all :( 22.620) 0.044 0.659 9.707 68.401 280.513 338.506 -> 103.401 -> 1654.414 MByte/s p45 cyclic-3dim-x :( 30.395) 0.033 0.479 7.290 49.467 185.663 216.171 -> 70.314 -> 1125.019 MByte/s p46 cyclic-3dim-y :( 16.887) 0.059 0.939 12.524 111.869 399.179 464.612 -> 162.229 -> 2595.663 MByte/s p47 cyclic-3dim-z :( 17.127) 0.058 0.925 12.834 111.765 408.952 466.163 -> 163.460 -> 2615.352 MByte/s p48 cyclic-3dim-all :( 22.764) 0.044 0.657 9.466 68.707 304.402 327.725 -> 104.228 -> 1667.650 MByte/s log_avg of all rings : 0.048 0.740 10.433 85.717 334.300 411.060 || 133.015 -> 2128.241 MByte/s log_avg of all random : 0.033 0.500 7.432 49.995 177.339 204.035 || 68.805 -> 1100.878 MByte/s log_avg(ring,random) : 0.040 0.608 8.806 65.463 243.484 289.604 || 95.667 -> 1530.664 MByte/s * size -> accumulated on all pr.: 0.637 9.735 140.897 1047.408 3895.748 4633.666 || 1530.664 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 1530.664 MByte/s on 16 processes ( = 95.667 MByte/s * 16 processes) Ping-pong latency: 11.811 microsec Ping-pong bandwidth: 969.781 MByte/s at Lmax= 8.000 MByte (MByte/s=1e6 Byte/s) (MByte=2**20 Byte) system parameters : 16 nodes, 1024 MB/node system name : HI-UX/MPP hostname : himiko OS release : 03-00 OS version : ad2b0 machine : SR8000 Date of measurement: Mon Nov 29 17:36:37 1999 Total execution wall clock time = 124 seconds SECTION-BEFF-END b_eff = 1530.664 MB/s = 95.667 * 16 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000