b_eff = 495.397 MB/s = 123.849 * 4 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 = 4 2-dim-paterns: size = 2 * 2 3-dim-paterns: size = 2 * 2 * 1 Used message lengths: msglng= 1 2 4 8 16 32 64 128 256 512 1024 2048 4096 10624 27554 71468 185364 480774 1246974 3234251 8388608 Used methods: 0=Sndrcv 1=Alltoal 2=non-blk Used patterns: 0=ring-2*2fix 1=ring-1*4fix 2=ring-1*4fix 3=ring-1*4fix 4=ring-1*4fix 5=ring-1*4fix 6=random-cyc-1dim 7=random-cyc-1dim 8=random-cyc-1dim 9=random-cyc-1dim 10=random-cyc-1dim 11=random-cyc-1dim 12=random-cyc-1dim 13=random-cyc-1dim 14=random-cyc-1dim 15=random-cyc-1dim 16=random-cyc-1dim 17=random-cyc-1dim 18=random-cyc-1dim 19=random-cyc-1dim 20=random-cyc-1dim 21=random-cyc-1dim 22=random-cyc-1dim 23=random-cyc-1dim 24=random-cyc-1dim 25=random-cyc-1dim 26=random-cyc-1dim 27=random-cyc-1dim 28=random-cyc-1dim 29=random-cyc-1dim 30=random-cyc-1dim 31=random-cyc-1dim 32=random-cyc-1dim 33=random-cyc-1dim 34=random-cyc-1dim 35=random-cyc-1dim 36=worst-cyc-1dim 37=best bi-section 38=worst bi-section 39=one PingPong Pair 40=acyclic-2dim-all 41=acyclic-3dim-all 42=cyclic-2dim-x 43=cyclic-2dim-y 44=cyclic-2dim-all 45=cyclic-3dim-x 46=cyclic-3dim-y 47=cyclic-3dim-all SECTION-HEAD-END SECTION-LOOP-BEGIN measurment loop: 85.804 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 7.0e-01 4.6e-03 2.1e-02 244 5.7e-01 3.7e-03 1.7e-02 244 5.6e-01 3.7e-03 1.7e-02 2 164 3.8e-01 2.5e-03 1.2e-02 164 3.8e-01 2.5e-03 1.2e-02 164 3.8e-01 2.5e-03 1.2e-02 4 161 3.8e-01 2.5e-03 1.2e-02 164 3.9e-01 2.5e-03 1.2e-02 163 3.8e-01 2.5e-03 1.2e-02 8 162 3.8e-01 2.5e-03 1.2e-02 163 3.8e-01 2.5e-03 1.2e-02 163 3.8e-01 2.5e-03 1.2e-02 16 163 4.0e-01 2.5e-03 1.2e-02 163 4.0e-01 2.5e-03 1.2e-02 163 4.0e-01 2.5e-03 1.2e-02 32 161 4.0e-01 2.5e-03 1.2e-02 162 4.0e-01 2.5e-03 1.2e-02 162 4.0e-01 2.5e-03 1.3e-02 64 160 4.1e-01 2.6e-03 1.3e-02 159 4.0e-01 2.6e-03 1.3e-02 159 4.0e-01 2.6e-03 1.3e-02 128 154 4.1e-01 2.7e-03 1.3e-02 154 4.1e-01 2.7e-03 1.3e-02 154 4.1e-01 2.7e-03 1.3e-02 256 145 3.9e-01 2.5e-03 1.2e-02 145 3.8e-01 2.5e-03 1.2e-02 144 3.9e-01 2.4e-03 1.2e-02 512 146 4.0e-01 2.6e-03 1.2e-02 147 3.9e-01 2.6e-03 1.2e-02 147 4.0e-01 2.5e-03 1.3e-02 1024 142 4.0e-01 2.6e-03 1.3e-02 143 4.0e-01 2.6e-03 1.3e-02 144 4.0e-01 2.7e-03 1.3e-02 2048 134 5.8e-01 3.3e-03 1.7e-02 135 5.8e-01 3.3e-03 1.7e-02 135 5.9e-01 3.3e-03 1.7e-02 4096 102 5.3e-01 3.2e-03 1.5e-02 102 5.3e-01 3.3e-03 1.5e-02 102 5.3e-01 3.3e-03 1.5e-02 10624 60 5.8e-01 3.1e-03 1.6e-02 59 5.7e-01 3.0e-03 1.6e-02 59 5.7e-01 3.0e-03 1.6e-02 27554 37 5.8e-01 3.0e-03 1.6e-02 38 5.9e-01 3.1e-03 1.7e-02 37 5.7e-01 3.0e-03 1.6e-02 71468 23 6.4e-01 3.6e-03 2.0e-02 23 6.4e-01 3.6e-03 2.0e-02 24 6.7e-01 3.7e-03 2.1e-02 185364 12 6.5e-01 4.2e-03 2.0e-02 12 6.5e-01 4.2e-03 2.0e-02 12 6.5e-01 4.2e-03 2.0e-02 480774 5 6.0e-01 5.4e-03 1.8e-02 5 6.0e-01 4.4e-03 1.9e-02 5 6.0e-01 4.6e-03 1.7e-02 1246974 1 2.9e-01 2.1e-03 1.1e-02 2 5.9e-01 4.3e-03 1.9e-02 2 5.6e-01 4.4e-03 1.6e-02 3234251 1 6.9e-01 6.0e-03 2.2e-02 1 6.8e-01 5.9e-03 2.0e-02 1 6.9e-01 5.9e-03 2.3e-02 8388608 1 1.7e+00 1.5e-02 4.9e-02 1 1.7e+00 1.5e-02 4.9e-02 1 1.7e+00 1.5e-02 4.9e-02 method=1 | ------- repetition 0 --------- ------- repetition 1 --------- ------- repetition 2 --------- [sec] (=Alltoal)| loop total per pattrn&mthd loop total per pattrn&mthd loop total per pattrn&mthd msg length| lng time minimum maximum lng time minimum maximum lng time minimum maximum ----------| ---- ------- ------- --------- ---- ------- ------- --------- ---- ------- ------- --------- 1 300 1.3e+00 2.5e-02 2.8e-02 43 1.8e-01 3.6e-03 4.1e-03 43 1.8e-01 3.6e-03 4.0e-03 2 150 6.3e-01 1.3e-02 1.4e-02 29 1.2e-01 2.5e-03 2.8e-03 29 1.2e-01 2.4e-03 2.8e-03 4 75 3.2e-01 6.4e-03 7.1e-03 29 1.2e-01 2.5e-03 2.8e-03 29 1.2e-01 2.4e-03 2.8e-03 8 37 1.6e-01 3.1e-03 3.5e-03 29 1.2e-01 2.4e-03 2.8e-03 29 1.2e-01 2.4e-03 2.8e-03 16 29 1.2e-01 2.5e-03 2.9e-03 29 1.2e-01 2.5e-03 2.8e-03 29 1.2e-01 2.5e-03 2.8e-03 32 29 1.3e-01 2.5e-03 2.9e-03 29 1.3e-01 2.5e-03 2.8e-03 29 1.3e-01 2.5e-03 2.9e-03 64 29 1.3e-01 2.5e-03 2.9e-03 29 1.3e-01 2.5e-03 2.9e-03 29 1.3e-01 2.5e-03 2.9e-03 128 29 1.3e-01 2.5e-03 3.0e-03 29 1.3e-01 2.5e-03 3.0e-03 29 1.3e-01 2.5e-03 3.0e-03 256 28 1.3e-01 2.5e-03 3.0e-03 28 1.3e-01 2.4e-03 2.9e-03 28 1.3e-01 2.5e-03 3.0e-03 512 28 1.3e-01 2.5e-03 3.0e-03 28 1.3e-01 2.5e-03 2.9e-03 28 1.3e-01 2.5e-03 3.0e-03 1024 28 1.3e-01 2.5e-03 3.1e-03 28 1.3e-01 2.5e-03 3.0e-03 28 1.3e-01 2.5e-03 3.0e-03 2048 27 1.6e-01 2.7e-03 4.0e-03 27 1.6e-01 2.7e-03 3.9e-03 28 1.7e-01 2.8e-03 4.1e-03 4096 25 1.7e-01 2.7e-03 4.3e-03 25 1.7e-01 2.7e-03 4.3e-03 24 1.7e-01 2.6e-03 4.1e-03 10624 17 1.8e-01 2.6e-03 4.3e-03 17 1.8e-01 2.6e-03 4.2e-03 17 1.8e-01 2.6e-03 4.3e-03 27554 12 1.9e-01 2.6e-03 4.8e-03 12 1.9e-01 2.7e-03 4.8e-03 12 1.9e-01 2.5e-03 4.7e-03 71468 8 2.2e-01 3.0e-03 5.8e-03 8 2.2e-01 2.9e-03 6.0e-03 9 2.4e-01 3.4e-03 6.5e-03 185364 5 2.5e-01 3.2e-03 7.3e-03 5 2.5e-01 3.2e-03 7.3e-03 5 2.5e-01 3.2e-03 7.0e-03 480774 2 2.4e-01 3.1e-03 6.7e-03 3 3.7e-01 4.6e-03 1.1e-02 3 3.6e-01 4.6e-03 1.0e-02 1246974 1 2.9e-01 3.5e-03 8.0e-03 1 2.9e-01 3.6e-03 8.2e-03 1 3.0e-01 3.6e-03 1.1e-02 3234251 1 7.5e-01 9.0e-03 2.1e-02 1 7.4e-01 9.1e-03 2.4e-02 1 7.4e-01 9.1e-03 2.1e-02 8388608 1 1.8e+00 2.2e-02 5.3e-02 1 1.8e+00 2.1e-02 5.3e-02 1 1.8e+00 2.2e-02 5.3e-02 method=2 | ------- repetition 0 --------- ------- repetition 1 --------- ------- repetition 2 --------- [sec] (=non-blk)| loop total per pattrn&mthd loop total per pattrn&mthd loop total per pattrn&mthd msg length| lng time minimum maximum lng time minimum maximum lng time minimum maximum ----------| ---- ------- ------- --------- ---- ------- ------- --------- ---- ------- ------- --------- 1 300 1.2e+00 1.1e-02 2.9e-02 103 4.1e-01 3.7e-03 9.8e-03 103 4.1e-01 3.7e-03 9.8e-03 2 150 6.1e-01 5.4e-03 1.5e-02 70 2.8e-01 2.5e-03 6.8e-03 70 2.8e-01 2.5e-03 6.7e-03 4 75 3.1e-01 2.7e-03 7.4e-03 69 2.8e-01 2.5e-03 6.7e-03 69 2.8e-01 2.5e-03 6.7e-03 8 69 2.8e-01 2.5e-03 6.7e-03 69 2.7e-01 2.5e-03 6.6e-03 69 2.7e-01 2.5e-03 6.6e-03 16 69 2.9e-01 2.5e-03 7.0e-03 69 2.9e-01 2.5e-03 6.9e-03 69 2.9e-01 2.5e-03 6.9e-03 32 69 2.9e-01 2.5e-03 7.1e-03 69 2.9e-01 2.5e-03 7.0e-03 69 2.9e-01 2.5e-03 7.0e-03 64 68 2.9e-01 2.5e-03 7.1e-03 69 2.9e-01 2.5e-03 7.1e-03 69 2.9e-01 2.5e-03 7.1e-03 128 68 3.0e-01 2.6e-03 7.2e-03 68 2.9e-01 2.5e-03 7.2e-03 68 2.9e-01 2.6e-03 7.2e-03 256 66 3.0e-01 2.5e-03 7.3e-03 66 2.9e-01 2.5e-03 7.0e-03 66 2.9e-01 2.5e-03 7.2e-03 512 65 3.0e-01 2.5e-03 7.3e-03 66 2.9e-01 2.5e-03 7.1e-03 66 2.9e-01 2.5e-03 7.3e-03 1024 64 3.0e-01 2.5e-03 7.3e-03 66 2.9e-01 2.6e-03 7.2e-03 65 2.9e-01 2.5e-03 7.2e-03 2048 62 3.6e-01 2.7e-03 9.0e-03 63 3.6e-01 2.7e-03 9.4e-03 64 3.6e-01 2.8e-03 9.2e-03 4096 57 3.8e-01 2.9e-03 9.8e-03 57 3.8e-01 2.9e-03 9.7e-03 56 3.7e-01 2.9e-03 9.6e-03 10624 37 3.3e-01 2.7e-03 1.1e-02 37 3.2e-01 2.6e-03 8.6e-03 37 3.2e-01 2.7e-03 8.5e-03 27554 26 3.5e-01 3.1e-03 1.1e-02 27 3.7e-01 2.7e-03 1.0e-02 26 3.5e-01 3.1e-03 9.5e-03 71468 16 4.3e-01 3.3e-03 1.5e-02 19 5.0e-01 3.7e-03 1.5e-02 15 4.0e-01 3.0e-03 1.2e-02 185364 9 4.7e-01 3.1e-03 1.8e-02 9 4.6e-01 3.6e-03 1.3e-02 9 4.6e-01 3.7e-03 1.3e-02 480774 5 6.0e-01 4.5e-03 1.8e-02 4 4.7e-01 3.5e-03 1.4e-02 4 4.8e-01 3.7e-03 1.4e-02 1246974 2 5.8e-01 4.2e-03 1.8e-02 2 5.6e-01 4.3e-03 2.0e-02 2 5.6e-01 4.3e-03 1.6e-02 3234251 1 6.9e-01 6.1e-03 2.1e-02 1 6.7e-01 6.0e-03 2.0e-02 1 6.8e-01 6.0e-03 2.0e-02 8388608 1 1.7e+00 1.5e-02 4.9e-02 1 1.7e+00 1.5e-02 4.9e-02 1 1.7e+00 1.5e-02 5.1e-02 SECTION-LOOPLNGS-END SECTION-ELAPSED-BEGIN measurment loop: elapsed time = 85.804 sec sum of max elapsed time per entries above = 85.378 sec difference to elapsed time = 0.427 sec = 0.5% sum based on fastest repetition = 79.782 sec difference to elapsed time = 6.023 sec = 7.0% 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-2*2fix 1 4 1.00 1.00 0 ( -1 -1 -1 ) p01 ring-1*4fix 2 8 2.00 1.00 0 ( -1 -1 -1 ) p02 ring-1*4fix 2 8 2.00 1.00 0 ( -1 -1 -1 ) p03 ring-1*4fix 2 8 2.00 1.00 0 ( -1 -1 -1 ) p04 ring-1*4fix 2 8 2.00 1.00 0 ( -1 -1 -1 ) p05 ring-1*4fix 2 8 2.00 1.00 0 ( -1 -1 -1 ) p06 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p07 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p08 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p09 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p10 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p11 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p12 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p13 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p14 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p15 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p16 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p17 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p18 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p19 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p20 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p21 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p22 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p23 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p24 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p25 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p26 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p27 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p28 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p29 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p30 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p31 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p32 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p33 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p34 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p35 random-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p36 worst-cyc-1dim 2 8 2.00 1.00 0 ( -1 -1 -1 ) p37 best bi-section 2 4 1.00 0.50 0 ( -1 -1 -1 ) p38 worst bi-section 2 4 1.00 0.50 0 ( -1 -1 -1 ) p39 one PingPong Pair 2 2 1.00 0.50 2 ( -1 -1 -1 ) p40 acyclic-2dim-all 4 8 2.00 0.50 0 ( -1 -1 -1 ) p41 acyclic-3dim-all 4 8 2.00 0.50 0 ( -1 -1 -1 ) p42 cyclic-2dim-x 1 4 1.00 1.00 0 ( -1 -1 -1 ) p43 cyclic-2dim-y 1 4 1.00 1.00 0 ( -1 -1 -1 ) p44 cyclic-2dim-all 2 8 2.00 1.00 0 ( -1 -1 -1 ) p45 cyclic-3dim-x 1 4 1.00 1.00 0 ( -1 -1 -1 ) p46 cyclic-3dim-y 1 4 1.00 1.00 0 ( -1 -1 -1 ) p47 cyclic-3dim-all 2 8 2.00 1.00 0 ( -1 -1 -1 ) SECTION-PATTERN-END Printing accumulated bandwidth / number of processes = bandwidth per process (in MBytes/sec = 1e6 Bytes/sec) Additionally the accumulated value processes is printed in the last column Only the cyclic and random patterns are used to compute b_eff (except p00), the other patterns are only as information. SECTION-BY-METHODS-BEGIN 2nd analysis path, only as information, last row: for all mthd: logavg_pat ( avg_L ( max_rep ( b(L) ) ) ) || logavg_pat ( max_mthd ( avg_L ( max_rep ( b(L) ) ) ) ) pattern / methods: Sndrcv Alltoal non-blk -> best method -> accumulated p00 ring-2*2fix : 116.423 101.104 108.849 -> 116.423 -> 465.693 MByte/s p01 ring-1*4fix : 102.457 109.180 105.195 -> 109.180 -> 436.720 MByte/s p02 ring-1*4fix : 102.097 109.124 104.893 -> 109.124 -> 436.495 MByte/s p03 ring-1*4fix : 102.762 109.243 107.683 -> 109.243 -> 436.973 MByte/s p04 ring-1*4fix : 102.945 109.328 104.553 -> 109.328 -> 437.313 MByte/s p05 ring-1*4fix : 103.187 109.338 107.190 -> 109.338 -> 437.353 MByte/s p06 random-cyc-1dim : 142.628 129.210 135.638 -> 142.628 -> 570.511 MByte/s p07 random-cyc-1dim : 103.690 108.962 105.167 -> 108.962 -> 435.849 MByte/s p08 random-cyc-1dim : 141.575 129.257 135.186 -> 141.575 -> 566.299 MByte/s p09 random-cyc-1dim : 142.393 129.424 133.907 -> 142.393 -> 569.573 MByte/s p10 random-cyc-1dim : 142.049 128.457 135.717 -> 142.049 -> 568.198 MByte/s p11 random-cyc-1dim : 103.003 109.385 105.260 -> 109.385 -> 437.541 MByte/s p12 random-cyc-1dim : 141.969 130.125 135.854 -> 141.969 -> 567.876 MByte/s p13 random-cyc-1dim : 141.969 130.179 134.697 -> 141.969 -> 567.874 MByte/s p14 random-cyc-1dim : 102.404 108.738 105.945 -> 108.738 -> 434.953 MByte/s p15 random-cyc-1dim : 142.690 128.546 134.282 -> 142.690 -> 570.759 MByte/s p16 random-cyc-1dim : 140.804 129.629 134.426 -> 140.804 -> 563.217 MByte/s p17 random-cyc-1dim : 101.456 108.984 106.499 -> 108.984 -> 435.937 MByte/s p18 random-cyc-1dim : 142.992 130.037 132.409 -> 142.992 -> 571.970 MByte/s p19 random-cyc-1dim : 102.346 109.276 106.034 -> 109.276 -> 437.103 MByte/s p20 random-cyc-1dim : 143.392 128.753 133.612 -> 143.392 -> 573.569 MByte/s p21 random-cyc-1dim : 143.244 129.876 132.758 -> 143.244 -> 572.976 MByte/s p22 random-cyc-1dim : 141.762 129.090 134.153 -> 141.762 -> 567.048 MByte/s p23 random-cyc-1dim : 142.749 128.492 135.682 -> 142.749 -> 570.995 MByte/s p24 random-cyc-1dim : 141.847 129.497 132.126 -> 141.847 -> 567.388 MByte/s p25 random-cyc-1dim : 143.040 130.079 134.315 -> 143.040 -> 572.161 MByte/s p26 random-cyc-1dim : 102.499 108.778 106.130 -> 108.778 -> 435.111 MByte/s p27 random-cyc-1dim : 142.650 128.863 133.158 -> 142.650 -> 570.599 MByte/s p28 random-cyc-1dim : 102.523 109.003 107.217 -> 109.003 -> 436.013 MByte/s p29 random-cyc-1dim : 143.095 129.993 134.106 -> 143.095 -> 572.382 MByte/s p30 random-cyc-1dim : 102.972 109.236 103.733 -> 109.236 -> 436.945 MByte/s p31 random-cyc-1dim : 142.854 128.826 135.753 -> 142.854 -> 571.414 MByte/s p32 random-cyc-1dim : 142.639 130.056 135.117 -> 142.639 -> 570.557 MByte/s p33 random-cyc-1dim : 142.362 130.624 134.605 -> 142.362 -> 569.450 MByte/s p34 random-cyc-1dim : 141.360 129.003 134.213 -> 141.360 -> 565.438 MByte/s p35 random-cyc-1dim : 142.315 129.536 136.615 -> 142.315 -> 569.260 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 141.760 127.837 135.288 -> 141.760 -> 567.039 MByte/s p37 best bi-section : 76.268 100.718 110.271 -> 110.271 -> 441.084 MByte/s p38 worst bi-section : 100.895 101.156 108.849 -> 108.849 -> 435.395 MByte/s p39 one PingPong Pair : 51.666 0.000 0.000 -> 51.666 -> 206.663 MByte/s p40 acyclic-2dim-all : 98.638 99.089 129.293 -> 129.293 -> 517.171 MByte/s p41 acyclic-3dim-all : 98.669 98.960 132.073 -> 132.073 -> 528.293 MByte/s p42 cyclic-2dim-x : 191.807 87.412 170.516 -> 191.807 -> 767.227 MByte/s p43 cyclic-2dim-y : 116.952 100.714 109.725 -> 116.952 -> 467.809 MByte/s p44 cyclic-2dim-all : 139.458 100.397 130.514 -> 139.458 -> 557.832 MByte/s p45 cyclic-3dim-x : 191.475 85.535 169.111 -> 191.475 -> 765.900 MByte/s p46 cyclic-3dim-y : 115.610 100.685 108.154 -> 115.610 -> 462.442 MByte/s p47 cyclic-3dim-all : 139.710 99.398 132.073 -> 139.710 -> 558.840 MByte/s log_avg of all rings : 104.860 107.842 106.382 || 110.408 -> 441.632 MByte/s log_avg of all random : 130.470 123.650 126.119 || 132.604 -> 530.416 MByte/s log_avg(ring,random) : 116.966 115.476 115.831 ||(120.998 -> 483.992)MByte/s * size -> accumulated on all pr.: 467.865 461.903 463.323 ||(483.992)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-2*2fix : 111.322 115.202 115.969 -> 115.969 -> 463.874 MByte/s p01 ring-1*4fix : 107.450 111.207 112.107 -> 112.107 -> 448.427 MByte/s p02 ring-1*4fix : 108.208 111.493 111.302 -> 111.493 -> 445.970 MByte/s p03 ring-1*4fix : 113.389 112.462 112.403 -> 113.389 -> 453.556 MByte/s p04 ring-1*4fix : 111.229 112.583 112.342 -> 112.583 -> 450.331 MByte/s p05 ring-1*4fix : 111.471 111.680 112.627 -> 112.627 -> 450.509 MByte/s p06 random-cyc-1dim : 142.129 143.634 136.741 -> 143.634 -> 574.538 MByte/s p07 random-cyc-1dim : 111.098 110.713 112.355 -> 112.355 -> 449.421 MByte/s p08 random-cyc-1dim : 142.378 140.833 141.086 -> 142.378 -> 569.511 MByte/s p09 random-cyc-1dim : 141.506 141.990 139.879 -> 141.990 -> 567.959 MByte/s p10 random-cyc-1dim : 141.574 140.945 142.354 -> 142.354 -> 569.417 MByte/s p11 random-cyc-1dim : 110.880 111.054 111.357 -> 111.357 -> 445.429 MByte/s p12 random-cyc-1dim : 140.779 140.985 142.200 -> 142.200 -> 568.801 MByte/s p13 random-cyc-1dim : 141.398 142.389 141.259 -> 142.389 -> 569.557 MByte/s p14 random-cyc-1dim : 111.252 111.122 112.156 -> 112.156 -> 448.626 MByte/s p15 random-cyc-1dim : 139.443 141.527 143.104 -> 143.104 -> 572.416 MByte/s p16 random-cyc-1dim : 141.849 140.806 137.007 -> 141.849 -> 567.398 MByte/s p17 random-cyc-1dim : 112.520 111.354 110.041 -> 112.520 -> 450.081 MByte/s p18 random-cyc-1dim : 142.696 141.502 141.797 -> 142.696 -> 570.785 MByte/s p19 random-cyc-1dim : 112.143 113.039 112.199 -> 113.039 -> 452.157 MByte/s p20 random-cyc-1dim : 141.029 141.119 140.978 -> 141.119 -> 564.476 MByte/s p21 random-cyc-1dim : 140.301 142.819 138.970 -> 142.819 -> 571.277 MByte/s p22 random-cyc-1dim : 137.948 138.905 141.725 -> 141.725 -> 566.900 MByte/s p23 random-cyc-1dim : 140.469 142.383 141.199 -> 142.383 -> 569.533 MByte/s p24 random-cyc-1dim : 142.132 140.521 140.424 -> 142.132 -> 568.530 MByte/s p25 random-cyc-1dim : 141.144 142.729 144.126 -> 144.126 -> 576.506 MByte/s p26 random-cyc-1dim : 110.902 109.930 111.259 -> 111.259 -> 445.038 MByte/s p27 random-cyc-1dim : 141.816 141.594 142.906 -> 142.906 -> 571.624 MByte/s p28 random-cyc-1dim : 111.780 112.042 111.777 -> 112.042 -> 448.170 MByte/s p29 random-cyc-1dim : 142.205 141.926 141.999 -> 142.205 -> 568.819 MByte/s p30 random-cyc-1dim : 110.861 112.005 111.179 -> 112.005 -> 448.019 MByte/s p31 random-cyc-1dim : 142.977 141.055 142.008 -> 142.977 -> 571.908 MByte/s p32 random-cyc-1dim : 141.343 141.855 142.734 -> 142.734 -> 570.936 MByte/s p33 random-cyc-1dim : 142.909 141.179 141.523 -> 142.909 -> 571.636 MByte/s p34 random-cyc-1dim : 141.522 141.537 141.428 -> 141.537 -> 566.147 MByte/s p35 random-cyc-1dim : 140.409 140.951 142.065 -> 142.065 -> 568.259 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 139.808 139.871 139.008 -> 139.871 -> 559.484 MByte/s p37 best bi-section : 109.617 107.587 106.682 -> 109.617 -> 438.466 MByte/s p38 worst bi-section : 107.785 107.861 109.690 -> 109.690 -> 438.759 MByte/s p39 one PingPong Pair : 51.392 51.252 51.023 -> 51.392 -> 205.567 MByte/s p40 acyclic-2dim-all : 126.264 128.362 125.831 -> 128.362 -> 513.446 MByte/s p41 acyclic-3dim-all : 125.615 129.133 128.242 -> 129.133 -> 516.533 MByte/s p42 cyclic-2dim-x : 188.947 186.693 187.679 -> 188.947 -> 755.788 MByte/s p43 cyclic-2dim-y : 116.287 115.636 116.769 -> 116.769 -> 467.076 MByte/s p44 cyclic-2dim-all : 138.783 137.400 137.899 -> 138.783 -> 555.133 MByte/s p45 cyclic-3dim-x : 188.492 188.741 189.222 -> 189.222 -> 756.888 MByte/s p46 cyclic-3dim-y : 115.541 115.426 116.059 -> 116.059 -> 464.238 MByte/s p47 cyclic-3dim-all : 140.167 141.097 138.443 -> 141.097 -> 564.387 MByte/s log_avg of all rings : 110.492 112.430 112.782 || 113.019 -> 452.075 MByte/s log_avg of all random : 132.668 132.762 132.622 || 133.639 -> 534.558 MByte/s log_avg(ring,random) : 121.073 122.174 122.300 ||(122.897 -> 491.589)MByte/s * size -> accumulated on all pr.: 484.293 488.695 489.201 ||(491.589)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-2*2fix p00 method 0 =Sndrcv :( 27.648) 0.036 0.568 8.357 62.854 325.626 382.761 -> 116.423 -> 465.693 MByte/s p00 method 1 =Alltoal :( 84.464) 0.012 0.188 2.908 33.075 288.455 376.914 -> 101.104 -> 404.415 MByte/s p00 method 2 =non-blk :( 51.060) 0.020 0.294 4.530 48.368 324.503 376.070 -> 108.849 -> 435.397 MByte/s p01 ring-1*4fix p01 method 0 =Sndrcv :( 27.609) 0.036 0.565 8.267 62.163 249.719 381.448 -> 102.457 -> 409.826 MByte/s p01 method 1 =Alltoal :( 42.454) 0.024 0.369 5.548 49.902 309.608 376.120 -> 109.180 -> 436.720 MByte/s p01 method 2 =non-blk :( 47.525) 0.021 0.321 4.840 48.554 276.249 364.484 -> 105.195 -> 420.780 MByte/s p02 ring-1*4fix p02 method 0 =Sndrcv :( 27.574) 0.036 0.564 8.347 62.680 245.637 378.983 -> 102.097 -> 408.389 MByte/s p02 method 1 =Alltoal :( 42.407) 0.024 0.371 5.412 49.866 312.326 381.334 -> 109.124 -> 436.495 MByte/s p02 method 2 =non-blk :( 47.427) 0.021 0.318 4.888 48.503 273.734 376.728 -> 104.893 -> 419.571 MByte/s p03 ring-1*4fix p03 method 0 =Sndrcv :( 27.650) 0.036 0.566 8.351 62.797 247.716 381.127 -> 102.762 -> 411.050 MByte/s p03 method 1 =Alltoal :( 42.359) 0.024 0.370 5.414 49.794 312.006 376.754 -> 109.243 -> 436.973 MByte/s p03 method 2 =non-blk :( 47.267) 0.021 0.322 4.936 48.705 275.795 410.482 -> 107.683 -> 430.733 MByte/s p04 ring-1*4fix p04 method 0 =Sndrcv :( 27.510) 0.036 0.569 8.280 62.930 235.920 380.349 -> 102.945 -> 411.780 MByte/s p04 method 1 =Alltoal :( 42.698) 0.023 0.367 5.529 49.824 313.541 379.267 -> 109.328 -> 437.313 MByte/s p04 method 2 =non-blk :( 47.427) 0.021 0.322 4.882 48.520 273.109 370.073 -> 104.553 -> 418.212 MByte/s p05 ring-1*4fix p05 method 0 =Sndrcv :( 27.428) 0.036 0.566 8.226 62.619 244.867 382.256 -> 103.187 -> 412.747 MByte/s p05 method 1 =Alltoal :( 42.888) 0.023 0.368 5.576 49.769 310.703 380.781 -> 109.338 -> 437.353 MByte/s p05 method 2 =non-blk :( 46.995) 0.021 0.325 4.834 48.463 276.664 406.543 -> 107.190 -> 428.759 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 24.303) 0.041 0.608 9.057 77.226 351.235 514.023 -> 142.628 -> 570.511 MByte/s p06 method 1 =Alltoal :( 45.850) 0.022 0.335 5.093 57.588 370.728 445.540 -> 129.210 -> 516.841 MByte/s p06 method 2 =non-blk :( 44.743) 0.022 0.347 5.200 56.268 352.849 498.890 -> 135.638 -> 542.554 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 27.463) 0.036 0.564 8.281 61.932 248.685 378.129 -> 103.690 -> 414.760 MByte/s p07 method 1 =Alltoal :( 42.418) 0.024 0.371 5.580 49.623 312.269 377.943 -> 108.962 -> 435.849 MByte/s p07 method 2 =non-blk :( 47.267) 0.021 0.319 4.816 48.412 275.225 362.774 -> 105.167 -> 420.667 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 24.334) 0.041 0.616 9.085 76.715 361.920 510.659 -> 141.575 -> 566.299 MByte/s p08 method 1 =Alltoal :( 46.012) 0.022 0.341 5.125 57.821 370.508 440.058 -> 129.257 -> 517.029 MByte/s p08 method 2 =non-blk :( 44.690) 0.022 0.344 5.224 56.395 350.735 496.236 -> 135.186 -> 540.743 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 24.162) 0.041 0.618 9.105 76.379 365.100 510.940 -> 142.393 -> 569.573 MByte/s p09 method 1 =Alltoal :( 46.291) 0.022 0.339 5.032 57.656 367.634 446.606 -> 129.424 -> 517.697 MByte/s p09 method 2 =non-blk :( 44.616) 0.022 0.343 5.326 56.306 340.847 491.109 -> 133.907 -> 535.626 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 24.295) 0.041 0.610 9.057 77.348 359.900 511.594 -> 142.049 -> 568.198 MByte/s p10 method 1 =Alltoal :( 45.801) 0.022 0.342 4.981 57.354 373.570 450.009 -> 128.457 -> 513.829 MByte/s p10 method 2 =non-blk :( 44.684) 0.022 0.345 5.253 56.199 350.003 494.626 -> 135.717 -> 542.869 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 27.592) 0.036 0.565 8.246 62.478 256.634 376.948 -> 103.003 -> 412.011 MByte/s p11 method 1 =Alltoal :( 42.663) 0.023 0.368 5.563 49.709 312.113 377.772 -> 109.385 -> 437.541 MByte/s p11 method 2 =non-blk :( 47.224) 0.021 0.323 4.827 48.393 279.421 383.147 -> 105.260 -> 421.042 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 24.281) 0.041 0.616 9.069 77.304 366.873 511.937 -> 141.969 -> 567.876 MByte/s p12 method 1 =Alltoal :( 46.070) 0.022 0.336 5.117 57.366 374.091 448.672 -> 130.125 -> 520.502 MByte/s p12 method 2 =non-blk :( 44.665) 0.022 0.346 5.211 55.796 342.738 492.882 -> 135.854 -> 543.417 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 24.266) 0.041 0.612 9.044 77.455 362.455 511.143 -> 141.969 -> 567.874 MByte/s p13 method 1 =Alltoal :( 45.814) 0.022 0.341 5.131 57.658 371.330 447.153 -> 130.179 -> 520.715 MByte/s p13 method 2 =non-blk :( 44.554) 0.022 0.342 5.237 56.538 342.039 482.186 -> 134.697 -> 538.788 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 27.615) 0.036 0.569 8.335 61.717 247.014 379.997 -> 102.404 -> 409.617 MByte/s p14 method 1 =Alltoal :( 42.907) 0.023 0.369 5.574 50.184 311.325 377.059 -> 108.738 -> 434.953 MByte/s p14 method 2 =non-blk :( 47.170) 0.021 0.322 4.885 48.623 274.050 370.660 -> 105.945 -> 423.779 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 24.252) 0.041 0.617 9.171 76.976 358.741 512.345 -> 142.690 -> 570.759 MByte/s p15 method 1 =Alltoal :( 46.210) 0.022 0.339 5.025 57.175 364.677 446.250 -> 128.546 -> 514.185 MByte/s p15 method 2 =non-blk :( 44.388) 0.023 0.343 5.304 57.037 344.150 499.455 -> 134.282 -> 537.127 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 24.313) 0.041 0.614 9.052 77.027 361.833 511.174 -> 140.804 -> 563.217 MByte/s p16 method 1 =Alltoal :( 45.896) 0.022 0.341 5.027 57.658 374.316 438.425 -> 129.629 -> 518.514 MByte/s p16 method 2 =non-blk :( 44.675) 0.022 0.343 5.303 55.559 342.596 489.561 -> 134.426 -> 537.704 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 27.425) 0.036 0.564 8.315 62.506 225.893 379.978 -> 101.456 -> 405.825 MByte/s p17 method 1 =Alltoal :( 42.710) 0.023 0.370 5.574 50.072 310.958 375.986 -> 108.984 -> 435.937 MByte/s p17 method 2 =non-blk :( 47.218) 0.021 0.321 4.843 48.675 270.384 412.612 -> 106.499 -> 425.994 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 24.344) 0.041 0.615 9.086 77.326 359.841 511.079 -> 142.992 -> 571.970 MByte/s p18 method 1 =Alltoal :( 45.836) 0.022 0.335 5.080 58.330 369.767 447.332 -> 130.037 -> 520.149 MByte/s p18 method 2 =non-blk :( 44.801) 0.022 0.347 5.237 56.151 343.656 494.248 -> 132.409 -> 529.636 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 27.617) 0.036 0.563 8.264 62.703 229.635 379.437 -> 102.346 -> 409.384 MByte/s p19 method 1 =Alltoal :( 42.337) 0.024 0.368 5.556 49.184 313.219 382.527 -> 109.276 -> 437.103 MByte/s p19 method 2 =non-blk :( 47.354) 0.021 0.321 4.854 48.358 275.225 409.820 -> 106.034 -> 424.134 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 24.268) 0.041 0.611 9.159 76.666 370.821 511.189 -> 143.392 -> 573.569 MByte/s p20 method 1 =Alltoal :( 46.338) 0.022 0.340 5.114 57.912 361.195 446.263 -> 128.753 -> 515.011 MByte/s p20 method 2 =non-blk :( 44.539) 0.022 0.344 5.268 56.035 342.039 472.743 -> 133.612 -> 534.448 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 24.158) 0.041 0.621 9.194 77.248 375.134 510.427 -> 143.244 -> 572.976 MByte/s p21 method 1 =Alltoal :( 45.930) 0.022 0.340 5.077 58.231 370.357 448.865 -> 129.876 -> 519.503 MByte/s p21 method 2 =non-blk :( 44.631) 0.022 0.342 5.306 56.254 330.976 484.500 -> 132.758 -> 531.034 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 24.131) 0.041 0.614 9.105 77.027 359.089 511.440 -> 141.762 -> 567.048 MByte/s p22 method 1 =Alltoal :( 45.792) 0.022 0.342 5.157 57.222 364.668 444.183 -> 129.090 -> 516.358 MByte/s p22 method 2 =non-blk :( 44.646) 0.022 0.342 5.228 55.359 352.254 489.631 -> 134.153 -> 536.611 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 24.205) 0.041 0.616 9.140 77.126 370.696 511.143 -> 142.749 -> 570.995 MByte/s p23 method 1 =Alltoal :( 45.919) 0.022 0.338 5.168 58.081 372.362 440.914 -> 128.492 -> 513.968 MByte/s p23 method 2 =non-blk :( 44.665) 0.022 0.346 5.243 56.218 350.038 488.150 -> 135.682 -> 542.729 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 24.168) 0.041 0.615 9.077 76.596 367.451 507.676 -> 141.847 -> 567.388 MByte/s p24 method 1 =Alltoal :( 46.483) 0.022 0.338 5.147 57.404 364.668 440.705 -> 129.497 -> 517.988 MByte/s p24 method 2 =non-blk :( 44.495) 0.022 0.347 5.223 56.136 336.108 447.321 -> 132.126 -> 528.505 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 24.281) 0.041 0.611 9.214 77.147 363.192 510.955 -> 143.040 -> 572.161 MByte/s p25 method 1 =Alltoal :( 45.803) 0.022 0.341 5.096 57.819 369.249 447.226 -> 130.079 -> 520.315 MByte/s p25 method 2 =non-blk :( 44.606) 0.022 0.345 5.334 56.489 334.893 482.257 -> 134.315 -> 537.259 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 27.541) 0.036 0.565 8.299 62.553 236.623 379.146 -> 102.499 -> 409.998 MByte/s p26 method 1 =Alltoal :( 42.660) 0.023 0.370 5.569 50.168 312.640 377.135 -> 108.778 -> 435.111 MByte/s p26 method 2 =non-blk :( 47.185) 0.021 0.320 4.900 48.358 276.317 374.282 -> 106.130 -> 424.518 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 24.336) 0.041 0.615 9.077 77.642 369.866 510.566 -> 142.650 -> 570.599 MByte/s p27 method 1 =Alltoal :( 46.431) 0.022 0.340 5.146 57.270 359.858 445.197 -> 128.863 -> 515.450 MByte/s p27 method 2 =non-blk :( 44.548) 0.022 0.344 5.210 55.715 333.223 499.693 -> 133.158 -> 532.630 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 27.597) 0.036 0.564 8.331 62.544 239.411 377.551 -> 102.523 -> 410.093 MByte/s p28 method 1 =Alltoal :( 42.383) 0.024 0.371 5.587 48.811 312.376 379.799 -> 109.003 -> 436.013 MByte/s p28 method 2 =non-blk :( 47.184) 0.021 0.319 4.801 48.665 274.591 395.801 -> 107.217 -> 428.867 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 24.117) 0.041 0.616 9.101 77.005 362.040 511.845 -> 143.095 -> 572.382 MByte/s p29 method 1 =Alltoal :( 46.205) 0.022 0.339 5.136 57.804 376.218 442.344 -> 129.993 -> 519.972 MByte/s p29 method 2 =non-blk :( 44.723) 0.022 0.344 5.214 55.681 339.669 491.728 -> 134.106 -> 536.423 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 27.395) 0.037 0.568 8.302 62.213 239.282 380.117 -> 102.972 -> 411.886 MByte/s p30 method 1 =Alltoal :( 42.419) 0.024 0.362 5.565 49.720 310.809 376.171 -> 109.236 -> 436.945 MByte/s p30 method 2 =non-blk :( 47.277) 0.021 0.323 4.853 48.929 277.331 354.331 -> 103.733 -> 414.931 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 24.212) 0.041 0.614 9.152 77.721 367.542 513.066 -> 142.854 -> 571.414 MByte/s p31 method 1 =Alltoal :( 45.954) 0.022 0.339 5.109 57.544 368.881 449.009 -> 128.826 -> 515.303 MByte/s p31 method 2 =non-blk :( 44.796) 0.022 0.347 5.303 55.768 341.022 492.130 -> 135.753 -> 543.012 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 24.256) 0.041 0.613 9.095 77.577 368.882 510.848 -> 142.639 -> 570.557 MByte/s p32 method 1 =Alltoal :( 46.500) 0.022 0.340 5.091 57.776 374.704 446.392 -> 130.056 -> 520.223 MByte/s p32 method 2 =non-blk :( 44.617) 0.022 0.343 5.281 56.261 361.685 492.723 -> 135.117 -> 540.466 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 24.389) 0.041 0.610 9.064 76.651 366.906 513.081 -> 142.362 -> 569.450 MByte/s p33 method 1 =Alltoal :( 46.116) 0.022 0.341 5.135 57.870 379.459 447.428 -> 130.624 -> 522.495 MByte/s p33 method 2 =non-blk :( 44.888) 0.022 0.343 5.245 56.393 341.159 487.129 -> 134.605 -> 538.421 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 24.191) 0.041 0.617 9.130 76.638 369.006 512.313 -> 141.360 -> 565.438 MByte/s p34 method 1 =Alltoal :( 46.272) 0.022 0.341 5.135 57.788 370.870 442.823 -> 129.003 -> 516.011 MByte/s p34 method 2 =non-blk :( 44.374) 0.023 0.343 5.238 56.896 348.612 490.605 -> 134.213 -> 536.853 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 24.238) 0.041 0.617 9.109 77.175 361.861 511.858 -> 142.315 -> 569.260 MByte/s p35 method 1 =Alltoal :( 46.162) 0.022 0.336 5.148 57.792 370.658 447.618 -> 129.536 -> 518.145 MByte/s p35 method 2 =non-blk :( 44.728) 0.022 0.346 5.274 55.807 351.546 493.056 -> 136.615 -> 546.461 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 24.199) 0.041 0.614 9.136 77.377 372.060 509.869 -> 141.760 -> 567.039 MByte/s p36 method 1 =Alltoal :( 46.478) 0.022 0.337 5.060 58.215 360.216 443.912 -> 127.837 -> 511.347 MByte/s p36 method 2 =non-blk :( 44.510) 0.022 0.345 5.310 56.455 353.526 488.819 -> 135.288 -> 541.153 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 24.066) 0.021 0.307 4.486 38.584 210.362 239.449 -> 76.268 -> 305.073 MByte/s p37 method 1 =Alltoal :( 42.558) 0.012 0.187 2.876 32.663 287.389 373.442 -> 100.718 -> 402.872 MByte/s p37 method 2 =non-blk :( 25.927) 0.019 0.295 4.598 48.077 308.310 376.779 -> 110.271 -> 441.084 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 23.873) 0.021 0.311 4.554 38.796 241.127 389.878 -> 100.895 -> 403.579 MByte/s p38 method 1 =Alltoal :( 42.361) 0.012 0.187 2.882 32.622 287.474 380.351 -> 101.156 -> 404.626 MByte/s p38 method 2 =non-blk :( 25.991) 0.019 0.297 4.634 47.509 316.446 378.359 -> 108.849 -> 435.395 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 23.160) 0.011 0.161 2.334 20.048 121.804 198.125 -> 51.666 -> 206.663 MByte/s p39 method 1 =Alltoal :(-------) 0.000 0.000 0.000 0.000 0.000 0.000 -> 0.000 -> 0.000 MByte/s p39 method 2 =non-blk :(-------) 0.000 0.000 0.000 0.000 0.000 0.000 -> 0.000 -> 0.000 MByte/s p40 acyclic-2dim-all p40 method 0 =Sndrcv :( 17.791) 0.028 0.427 6.221 54.843 248.158 343.275 -> 98.638 -> 394.554 MByte/s p40 method 1 =Alltoal :( 21.622) 0.023 0.357 5.394 56.002 270.172 316.891 -> 99.089 -> 396.356 MByte/s p40 method 2 =non-blk :( 21.211) 0.024 0.366 5.593 58.434 342.810 430.009 -> 129.293 -> 517.171 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 17.817) 0.028 0.426 6.190 55.513 249.467 341.202 -> 98.669 -> 394.675 MByte/s p41 method 1 =Alltoal :( 21.663) 0.023 0.363 5.404 54.342 282.225 318.046 -> 98.960 -> 395.838 MByte/s p41 method 2 =non-blk :( 21.235) 0.024 0.367 5.604 58.427 325.231 467.346 -> 132.073 -> 528.293 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 15.188) 0.066 1.018 15.034 129.067 501.771 549.172 -> 191.807 -> 767.227 MByte/s p42 method 1 =Alltoal :( 84.780) 0.012 0.189 2.925 37.385 250.563 273.691 -> 87.412 -> 349.650 MByte/s p42 method 2 =non-blk :( 35.495) 0.028 0.447 6.786 80.037 463.287 547.378 -> 170.516 -> 682.063 MByte/s p43 cyclic-2dim-y p43 method 0 =Sndrcv :( 27.725) 0.036 0.563 8.227 63.226 322.698 378.821 -> 116.952 -> 467.809 MByte/s p43 method 1 =Alltoal :( 84.325) 0.012 0.186 2.904 33.129 288.455 374.575 -> 100.714 -> 402.857 MByte/s p43 method 2 =non-blk :( 52.017) 0.019 0.295 4.684 48.443 323.685 371.606 -> 109.725 -> 438.902 MByte/s p44 cyclic-2dim-all p44 method 0 =Sndrcv :( 21.395) 0.047 0.728 10.661 83.433 381.017 450.106 -> 139.458 -> 557.832 MByte/s p44 method 1 =Alltoal :( 42.710) 0.023 0.372 5.570 56.340 280.388 319.262 -> 100.397 -> 401.587 MByte/s p44 method 2 =non-blk :( 42.913) 0.023 0.355 5.473 57.378 321.876 446.999 -> 130.514 -> 522.055 MByte/s p45 cyclic-3dim-x p45 method 0 =Sndrcv :( 15.207) 0.066 1.036 15.102 126.642 504.171 548.526 -> 191.475 -> 765.900 MByte/s p45 method 1 =Alltoal :( 85.326) 0.012 0.189 2.942 36.887 235.057 273.664 -> 85.535 -> 342.138 MByte/s p45 method 2 =non-blk :( 35.447) 0.028 0.446 6.764 79.546 447.138 548.920 -> 169.111 -> 676.445 MByte/s p46 cyclic-3dim-y p46 method 0 =Sndrcv :( 27.636) 0.036 0.569 8.390 61.703 331.257 366.586 -> 115.610 -> 462.442 MByte/s p46 method 1 =Alltoal :( 83.746) 0.012 0.189 2.924 33.009 288.734 379.953 -> 100.685 -> 402.741 MByte/s p46 method 2 =non-blk :( 51.893) 0.019 0.293 4.571 46.658 275.521 380.470 -> 108.154 -> 432.617 MByte/s p47 cyclic-3dim-all p47 method 0 =Sndrcv :( 21.370) 0.047 0.736 10.844 83.433 377.204 450.201 -> 139.710 -> 558.840 MByte/s p47 method 1 =Alltoal :( 42.768) 0.023 0.370 5.426 56.623 277.992 317.955 -> 99.398 -> 397.592 MByte/s p47 method 2 =non-blk :( 43.160) 0.023 0.360 5.579 57.216 344.366 492.549 -> 132.073 -> 528.291 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.036 0.566 8.305 62.673 256.656 381.152 || 104.860 -> 419.440 MByte/s - ring, method 1 = Alltoal: 0.021 0.330 4.942 46.541 307.645 378.523 || 107.842 -> 431.368 MByte/s - ring, method 2 = non-blk: 0.021 0.317 4.817 48.519 282.783 383.658 || 106.382 -> 425.527 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.040 0.601 8.883 72.842 326.156 472.115 || 130.470 -> 521.881 MByte/s - random, method 1 = Alltoal: 0.022 0.347 5.224 55.434 353.515 426.344 || 123.650 -> 494.600 MByte/s - random, method 2 = non-blk: 0.022 0.338 5.145 54.012 324.261 457.656 || 126.119 -> 504.475 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.038 0.583 8.589 67.567 289.327 424.202 || 116.966 -> 467.865 MByte/s - average, method 1 = Alltoal: 0.022 0.338 5.081 50.793 329.784 401.722 || 115.476 -> 461.903 MByte/s - average, method 2 = non-blk: 0.021 0.327 4.978 51.192 302.813 419.026 || 115.831 -> 463.323 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 0.152 2.333 34.356 270.266 1157.307 1696.809 || 467.865 MByte/s - accumulated, mthd 1 = Alltoal: 0.086 1.353 20.324 203.172 1319.134 1606.890 || 461.903 MByte/s - accumulated, mthd 2 = non-blk: 0.086 1.309 19.912 204.767 1211.251 1676.106 || 463.323 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 0.152 0.038 0.036 0.040 0.038 0.022 0.021 2 0.302 0.076 0.072 0.079 0.076 0.043 0.043 4 0.601 0.150 0.144 0.157 0.150 0.086 0.085 8 1.211 0.303 0.289 0.317 0.303 0.172 0.171 16 2.333 0.583 0.566 0.601 0.583 0.338 0.327 32 4.610 1.153 1.121 1.185 1.153 0.670 0.650 64 9.020 2.255 2.193 2.319 2.255 1.327 1.282 128 17.323 4.331 4.216 4.449 4.331 2.593 2.518 256 34.356 8.589 8.305 8.883 8.589 5.081 4.978 512 67.597 16.899 16.263 17.560 16.899 10.063 9.859 1024 131.812 32.953 31.870 34.073 32.953 19.869 19.341 2048 163.932 40.983 38.315 43.837 40.983 29.623 30.117 4096 270.266 67.567 62.673 72.842 67.567 50.793 51.192 10624 404.269 101.067 94.479 108.115 92.333 89.318 100.283 27554 681.587 170.397 157.292 184.594 147.768 151.190 166.518 71468 948.792 237.198 221.334 254.199 217.539 227.835 218.582 185364 1334.174 333.543 313.924 354.390 289.327 329.784 302.813 480774 1421.612 355.403 326.635 386.704 340.449 345.361 334.374 1246974 1547.618 386.905 350.734 426.806 367.774 376.732 367.854 3234251 1633.783 408.446 368.317 452.947 398.646 380.824 400.749 8388608 1723.397 430.849 390.279 475.636 424.202 401.722 419.026 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-2*2fix :( 27.648) 0.036 0.568 8.357 62.854 325.626 382.761 -> 116.935 -> 467.740 MByte/s p01 ring-1*4fix :( 27.609) 0.036 0.565 8.267 62.163 309.608 381.448 -> 112.871 -> 451.483 MByte/s p02 ring-1*4fix :( 27.574) 0.036 0.564 8.347 62.680 312.326 381.334 -> 111.935 -> 447.739 MByte/s p03 ring-1*4fix :( 27.650) 0.036 0.566 8.351 62.797 312.006 410.482 -> 114.272 -> 457.090 MByte/s p04 ring-1*4fix :( 27.510) 0.036 0.569 8.280 62.930 313.541 380.349 -> 113.145 -> 452.581 MByte/s p05 ring-1*4fix :( 27.428) 0.036 0.566 8.226 62.619 310.703 406.543 -> 113.708 -> 454.830 MByte/s p06 random-cyc-1dim :( 24.303) 0.041 0.608 9.057 77.226 370.728 514.023 -> 144.215 -> 576.861 MByte/s p07 random-cyc-1dim :( 27.463) 0.036 0.564 8.281 61.932 312.269 378.129 -> 112.752 -> 451.007 MByte/s p08 random-cyc-1dim :( 24.334) 0.041 0.616 9.085 76.715 370.508 510.659 -> 143.186 -> 572.743 MByte/s p09 random-cyc-1dim :( 24.162) 0.041 0.618 9.105 76.379 367.634 510.940 -> 143.099 -> 572.395 MByte/s p10 random-cyc-1dim :( 24.295) 0.041 0.610 9.057 77.348 373.570 511.594 -> 143.640 -> 574.560 MByte/s p11 random-cyc-1dim :( 27.592) 0.036 0.565 8.246 62.478 312.113 383.147 -> 112.192 -> 448.767 MByte/s p12 random-cyc-1dim :( 24.281) 0.041 0.616 9.069 77.304 374.091 511.937 -> 143.403 -> 573.611 MByte/s p13 random-cyc-1dim :( 24.266) 0.041 0.612 9.044 77.455 371.330 511.143 -> 143.672 -> 574.690 MByte/s p14 random-cyc-1dim :( 27.615) 0.036 0.569 8.335 61.717 311.325 379.997 -> 112.824 -> 451.294 MByte/s p15 random-cyc-1dim :( 24.252) 0.041 0.617 9.171 76.976 364.677 512.345 -> 143.918 -> 575.674 MByte/s p16 random-cyc-1dim :( 24.313) 0.041 0.614 9.052 77.027 374.316 511.174 -> 143.024 -> 572.095 MByte/s p17 random-cyc-1dim :( 27.425) 0.036 0.564 8.315 62.506 310.958 412.612 -> 113.560 -> 454.240 MByte/s p18 random-cyc-1dim :( 24.344) 0.041 0.615 9.086 77.326 369.767 511.079 -> 144.209 -> 576.834 MByte/s p19 random-cyc-1dim :( 27.617) 0.036 0.563 8.264 62.703 313.219 409.820 -> 113.988 -> 455.950 MByte/s p20 random-cyc-1dim :( 24.268) 0.041 0.611 9.159 76.666 370.821 511.189 -> 144.601 -> 578.404 MByte/s p21 random-cyc-1dim :( 24.158) 0.041 0.621 9.194 77.248 375.134 510.427 -> 144.246 -> 576.985 MByte/s p22 random-cyc-1dim :( 24.131) 0.041 0.614 9.105 77.027 364.668 511.440 -> 143.024 -> 572.096 MByte/s p23 random-cyc-1dim :( 24.205) 0.041 0.616 9.140 77.126 372.362 511.143 -> 143.949 -> 575.797 MByte/s p24 random-cyc-1dim :( 24.168) 0.041 0.615 9.077 76.596 367.451 507.676 -> 143.040 -> 572.162 MByte/s p25 random-cyc-1dim :( 24.281) 0.041 0.611 9.214 77.147 369.249 510.955 -> 144.946 -> 579.786 MByte/s p26 random-cyc-1dim :( 27.541) 0.036 0.565 8.299 62.553 312.640 379.146 -> 111.642 -> 446.569 MByte/s p27 random-cyc-1dim :( 24.336) 0.041 0.615 9.077 77.642 369.866 510.566 -> 143.838 -> 575.350 MByte/s p28 random-cyc-1dim :( 27.597) 0.036 0.564 8.331 62.544 312.376 395.801 -> 113.577 -> 454.309 MByte/s p29 random-cyc-1dim :( 24.117) 0.041 0.616 9.101 77.005 376.218 511.845 -> 144.860 -> 579.442 MByte/s p30 random-cyc-1dim :( 27.395) 0.037 0.568 8.302 62.213 310.809 380.117 -> 112.515 -> 450.058 MByte/s p31 random-cyc-1dim :( 24.212) 0.041 0.614 9.152 77.721 368.881 513.066 -> 143.983 -> 575.932 MByte/s p32 random-cyc-1dim :( 24.256) 0.041 0.613 9.095 77.577 374.704 510.848 -> 143.960 -> 575.838 MByte/s p33 random-cyc-1dim :( 24.389) 0.041 0.610 9.064 76.651 379.459 513.081 -> 144.020 -> 576.081 MByte/s p34 random-cyc-1dim :( 24.191) 0.041 0.617 9.130 76.638 370.870 512.313 -> 142.196 -> 568.786 MByte/s p35 random-cyc-1dim :( 24.238) 0.041 0.617 9.109 77.175 370.658 511.858 -> 143.848 -> 575.390 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 24.199) 0.041 0.614 9.136 77.377 372.060 509.869 -> 142.557 -> 570.228 MByte/s p37 best bi-section :( 24.066) 0.021 0.307 4.598 48.077 308.310 376.779 -> 110.274 -> 441.095 MByte/s p38 worst bi-section :( 23.873) 0.021 0.311 4.634 47.509 316.446 389.878 -> 111.560 -> 446.241 MByte/s p39 one PingPong Pair :( 23.160) 0.011 0.161 2.334 20.048 121.804 198.125 -> 51.666 -> 206.663 MByte/s p40 acyclic-2dim-all :( 17.791) 0.028 0.427 6.221 58.434 342.810 430.009 -> 129.491 -> 517.966 MByte/s p41 acyclic-3dim-all :( 17.817) 0.028 0.426 6.190 58.427 325.231 467.346 -> 132.280 -> 529.122 MByte/s p42 cyclic-2dim-x :( 15.188) 0.066 1.018 15.034 129.067 501.771 549.172 -> 192.117 -> 768.469 MByte/s p43 cyclic-2dim-y :( 27.725) 0.036 0.563 8.227 63.226 323.685 378.821 -> 117.863 -> 471.453 MByte/s p44 cyclic-2dim-all :( 21.395) 0.047 0.728 10.661 83.433 381.017 450.106 -> 141.555 -> 566.218 MByte/s p45 cyclic-3dim-x :( 15.207) 0.066 1.036 15.102 126.642 504.171 548.920 -> 191.825 -> 767.299 MByte/s p46 cyclic-3dim-y :( 27.636) 0.036 0.569 8.390 61.703 331.257 380.470 -> 117.081 -> 468.325 MByte/s p47 cyclic-3dim-all :( 21.370) 0.047 0.736 10.844 83.433 377.204 492.549 -> 141.726 -> 566.906 MByte/s log_avg of all rings : 0.036 0.566 8.305 62.673 313.924 390.279 || 113.800 -> 455.201 MByte/s log_avg of all random : 0.040 0.601 8.883 72.842 354.390 475.636 || 134.786 -> 539.143 MByte/s log_avg(ring,random) : 0.038 0.583 8.589 67.567 333.543 430.849 || 123.849 -> 495.397 MByte/s * size -> accumulated on all pr.: 0.152 2.333 34.356 270.266 1334.174 1723.397 || 495.397 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 495.397 MByte/s on 4 processes ( = 123.849 MByte/s * 4 processes) Ping-pong latency: 23.160 microsec Ping-pong bandwidth: 792.499 MByte/s at Lmax= 8.000 MByte (MByte/s=1e6 Byte/s) (MByte=2**20 Byte) system parameters : 4 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 14:23:10 1999 Total execution wall clock time = 87 seconds SECTION-BEFF-END b_eff = 495.397 MB/s = 123.849 * 4 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000