b_eff = 429.108 MB/s = 143.036 * 3 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 = 3 2-dim-paterns: size = 3 * 1 3-dim-paterns: size = 3 * 1 * 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-1*3fix 1=ring-1*3fix 2=ring-1*3fix 3=ring-1*3fix 4=ring-1*3fix 5=ring-1*3fix 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-all 44=cyclic-3dim-x 45=cyclic-3dim-all SECTION-HEAD-END SECTION-LOOP-BEGIN measurment loop: 63.931 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 7.2e-01 1.4e-02 2.1e-02 81 1.9e-01 3.7e-03 4.3e-03 81 1.9e-01 3.7e-03 4.3e-03 2 150 3.6e-01 6.9e-03 7.9e-03 54 1.3e-01 2.5e-03 2.9e-03 54 1.3e-01 2.5e-03 2.9e-03 4 75 1.8e-01 3.5e-03 4.1e-03 54 1.3e-01 2.5e-03 2.9e-03 54 1.3e-01 2.5e-03 2.9e-03 8 53 1.3e-01 2.4e-03 2.9e-03 53 1.3e-01 2.4e-03 2.9e-03 53 1.3e-01 2.4e-03 2.8e-03 16 54 1.3e-01 2.7e-03 3.2e-03 54 1.3e-01 2.7e-03 3.0e-03 54 1.3e-01 2.7e-03 3.0e-03 32 50 1.2e-01 2.5e-03 2.8e-03 50 1.2e-01 2.5e-03 2.8e-03 50 1.2e-01 2.5e-03 2.8e-03 64 49 1.2e-01 2.5e-03 2.8e-03 49 1.2e-01 2.5e-03 2.8e-03 49 1.2e-01 2.5e-03 2.8e-03 128 48 1.3e-01 2.6e-03 2.9e-03 48 1.3e-01 2.6e-03 2.9e-03 48 1.3e-01 2.6e-03 2.9e-03 256 45 1.2e-01 2.4e-03 2.7e-03 46 1.2e-01 2.5e-03 2.8e-03 45 1.2e-01 2.4e-03 2.8e-03 512 46 1.3e-01 2.5e-03 2.9e-03 46 1.3e-01 2.5e-03 2.9e-03 46 1.3e-01 2.5e-03 2.9e-03 1024 45 1.3e-01 2.5e-03 2.9e-03 45 1.3e-01 2.5e-03 2.9e-03 45 1.3e-01 2.5e-03 2.9e-03 2048 44 2.0e-01 3.8e-03 4.5e-03 44 2.0e-01 3.7e-03 4.5e-03 44 2.0e-01 3.7e-03 4.6e-03 4096 29 1.6e-01 2.9e-03 3.7e-03 29 1.7e-01 2.9e-03 3.7e-03 29 1.6e-01 2.9e-03 3.7e-03 10624 19 1.6e-01 2.9e-03 3.8e-03 19 1.6e-01 2.9e-03 3.9e-03 19 1.6e-01 2.9e-03 3.7e-03 27554 12 1.7e-01 3.0e-03 4.6e-03 12 1.7e-01 3.0e-03 4.9e-03 12 1.7e-01 3.0e-03 4.6e-03 71468 7 1.7e-01 2.8e-03 4.8e-03 7 1.7e-01 2.8e-03 4.5e-03 7 1.7e-01 2.8e-03 4.7e-03 185364 4 2.0e-01 3.0e-03 5.9e-03 4 2.0e-01 3.0e-03 5.9e-03 4 1.9e-01 3.0e-03 5.7e-03 480774 2 2.3e-01 3.0e-03 6.5e-03 2 2.3e-01 2.9e-03 6.0e-03 2 2.3e-01 2.9e-03 6.4e-03 1246974 1 2.6e-01 3.4e-03 1.1e-02 1 2.5e-01 3.4e-03 7.7e-03 1 2.4e-01 3.4e-03 7.7e-03 3234251 1 6.1e-01 8.4e-03 1.8e-02 1 6.0e-01 8.3e-03 1.5e-02 1 5.9e-01 8.3e-03 1.5e-02 8388608 1 1.6e+00 2.1e-02 3.7e-02 1 1.6e+00 2.1e-02 3.8e-02 1 1.6e+00 2.1e-02 3.7e-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 9.8e-01 2.1e-02 2.2e-02 52 1.7e-01 3.7e-03 3.8e-03 52 1.7e-01 3.7e-03 3.8e-03 2 150 4.9e-01 1.1e-02 1.1e-02 35 1.1e-01 2.5e-03 2.6e-03 34 1.1e-01 2.4e-03 2.6e-03 4 75 2.5e-01 5.3e-03 5.6e-03 35 1.2e-01 2.5e-03 2.7e-03 35 1.2e-01 2.5e-03 2.6e-03 8 37 1.2e-01 2.6e-03 2.8e-03 34 1.1e-01 2.4e-03 2.5e-03 35 1.2e-01 2.5e-03 2.6e-03 16 35 1.2e-01 2.5e-03 2.8e-03 35 1.2e-01 2.5e-03 2.8e-03 35 1.2e-01 2.5e-03 2.8e-03 32 34 1.2e-01 2.5e-03 2.8e-03 34 1.2e-01 2.5e-03 2.8e-03 34 1.2e-01 2.5e-03 2.8e-03 64 34 1.2e-01 2.5e-03 2.8e-03 34 1.2e-01 2.5e-03 2.8e-03 33 1.2e-01 2.4e-03 2.7e-03 128 34 1.3e-01 2.5e-03 3.9e-03 33 1.2e-01 2.4e-03 2.8e-03 33 1.2e-01 2.4e-03 2.9e-03 256 33 1.3e-01 2.5e-03 2.9e-03 33 1.3e-01 2.5e-03 2.9e-03 33 1.3e-01 2.5e-03 2.9e-03 512 33 1.3e-01 2.5e-03 2.9e-03 33 1.3e-01 2.5e-03 3.0e-03 33 1.3e-01 2.5e-03 3.0e-03 1024 33 1.3e-01 2.6e-03 3.7e-03 33 1.3e-01 2.5e-03 3.0e-03 32 1.3e-01 2.5e-03 2.9e-03 2048 32 1.9e-01 3.0e-03 4.3e-03 33 1.9e-01 3.2e-03 4.4e-03 32 1.9e-01 3.2e-03 4.3e-03 4096 26 1.8e-01 2.8e-03 4.2e-03 26 1.8e-01 2.8e-03 4.2e-03 25 1.7e-01 2.7e-03 4.0e-03 10624 17 2.0e-01 2.3e-03 4.6e-03 17 2.0e-01 2.3e-03 4.6e-03 18 2.1e-01 2.4e-03 4.9e-03 27554 14 2.5e-01 2.5e-03 5.9e-03 14 2.6e-01 2.5e-03 6.1e-03 14 2.5e-01 2.6e-03 6.0e-03 71468 10 3.2e-01 2.9e-03 7.5e-03 10 3.1e-01 3.0e-03 7.5e-03 10 3.2e-01 2.9e-03 7.6e-03 185364 6 3.4e-01 2.9e-03 8.5e-03 6 3.4e-01 2.9e-03 8.5e-03 6 3.4e-01 2.9e-03 8.4e-03 480774 4 5.4e-01 4.4e-03 1.3e-02 4 5.5e-01 4.4e-03 1.4e-02 3 4.1e-01 3.3e-03 1.1e-02 1246974 1 3.2e-01 2.6e-03 1.1e-02 1 3.2e-01 2.6e-03 1.0e-02 1 3.2e-01 2.6e-03 1.0e-02 3234251 1 7.9e-01 6.4e-03 2.3e-02 1 7.8e-01 6.4e-03 1.9e-02 1 7.8e-01 6.4e-03 1.9e-02 8388608 1 2.1e+00 1.7e-02 5.2e-02 1 2.1e+00 1.7e-02 5.2e-02 1 2.1e+00 1.7e-02 5.2e-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 1.5e-02 2.8e-02 73 3.0e-01 3.6e-03 6.8e-03 75 3.0e-01 3.7e-03 6.9e-03 2 150 6.1e-01 7.7e-03 1.4e-02 50 2.0e-01 2.5e-03 4.7e-03 50 2.0e-01 2.5e-03 4.7e-03 4 75 3.1e-01 3.8e-03 7.1e-03 50 2.0e-01 2.5e-03 4.7e-03 50 2.0e-01 2.5e-03 4.7e-03 8 48 2.0e-01 2.4e-03 4.5e-03 49 2.0e-01 2.4e-03 4.5e-03 49 2.0e-01 2.4e-03 4.6e-03 16 49 2.1e-01 2.6e-03 4.8e-03 50 2.1e-01 2.6e-03 4.9e-03 50 2.1e-01 2.6e-03 4.9e-03 32 47 2.0e-01 2.5e-03 4.7e-03 48 2.0e-01 2.5e-03 4.7e-03 47 2.0e-01 2.5e-03 4.6e-03 64 46 2.0e-01 2.5e-03 4.6e-03 47 2.0e-01 2.5e-03 4.7e-03 47 2.0e-01 2.5e-03 4.7e-03 128 46 2.1e-01 2.6e-03 4.7e-03 47 2.1e-01 2.6e-03 4.8e-03 46 2.0e-01 2.6e-03 4.7e-03 256 45 2.0e-01 2.4e-03 4.7e-03 45 2.0e-01 2.4e-03 4.7e-03 44 2.0e-01 2.3e-03 4.6e-03 512 46 2.1e-01 2.5e-03 4.9e-03 47 2.1e-01 2.5e-03 4.9e-03 47 2.1e-01 2.5e-03 4.9e-03 1024 45 2.1e-01 2.5e-03 4.8e-03 46 2.1e-01 2.6e-03 4.9e-03 46 2.1e-01 2.5e-03 4.9e-03 2048 45 2.7e-01 3.2e-03 6.3e-03 45 2.7e-01 3.2e-03 6.2e-03 46 2.8e-01 3.2e-03 6.4e-03 4096 34 2.4e-01 2.8e-03 5.6e-03 35 2.5e-01 2.9e-03 5.7e-03 35 2.5e-01 2.9e-03 5.7e-03 10624 23 2.2e-01 2.6e-03 7.7e-03 23 2.1e-01 2.5e-03 5.0e-03 23 2.1e-01 2.5e-03 5.0e-03 27554 17 2.5e-01 3.1e-03 7.0e-03 17 2.4e-01 2.6e-03 5.8e-03 17 2.5e-01 3.1e-03 5.9e-03 71468 10 3.2e-01 2.8e-03 1.1e-02 12 3.8e-01 3.1e-03 9.2e-03 10 3.1e-01 2.6e-03 7.6e-03 185364 6 3.6e-01 2.8e-03 1.3e-02 7 4.1e-01 3.2e-03 1.0e-02 7 4.1e-01 3.1e-03 1.0e-02 480774 4 5.4e-01 4.3e-03 1.6e-02 4 5.5e-01 4.3e-03 1.4e-02 4 5.4e-01 4.2e-03 1.4e-02 1246974 1 2.6e-01 2.5e-03 8.2e-03 1 2.6e-01 2.5e-03 7.9e-03 1 2.6e-01 2.5e-03 7.9e-03 3234251 1 6.4e-01 6.4e-03 1.7e-02 1 6.3e-01 6.4e-03 1.6e-02 1 6.4e-01 6.4e-03 1.6e-02 8388608 1 1.7e+00 1.7e-02 4.0e-02 1 1.7e+00 1.7e-02 4.0e-02 1 1.7e+00 1.7e-02 4.0e-02 SECTION-LOOPLNGS-END SECTION-ELAPSED-BEGIN measurment loop: elapsed time = 63.931 sec sum of max elapsed time per entries above = 63.897 sec difference to elapsed time = 0.035 sec = 0.1% sum based on fastest repetition = 58.500 sec difference to elapsed time = 5.431 sec = 8.5% 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-1*3fix 2 6 2.00 1.00 0 ( -1 -1 -1 ) p01 ring-1*3fix 2 6 2.00 1.00 0 ( -1 -1 -1 ) p02 ring-1*3fix 2 6 2.00 1.00 0 ( -1 -1 -1 ) p03 ring-1*3fix 2 6 2.00 1.00 0 ( -1 -1 -1 ) p04 ring-1*3fix 2 6 2.00 1.00 0 ( -1 -1 -1 ) p05 ring-1*3fix 2 6 2.00 1.00 0 ( -1 -1 -1 ) p06 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p07 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p08 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p09 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p10 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p11 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p12 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p13 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p14 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p15 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p16 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p17 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p18 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p19 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p20 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p21 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p22 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p23 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p24 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p25 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p26 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p27 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p28 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p29 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p30 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p31 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p32 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p33 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p34 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p35 random-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p36 worst-cyc-1dim 2 6 2.00 1.00 0 ( -1 -1 -1 ) p37 best bi-section 2 2 1.00 0.50 1 ( -1 -1 -1 ) p38 worst bi-section 2 2 1.00 0.50 1 ( -1 -1 -1 ) p39 one PingPong Pair 2 2 1.00 0.50 1 ( -1 -1 -1 ) p40 acyclic-2dim-all 2 4 1.33 0.67 0 ( -1 -1 -1 ) p41 acyclic-3dim-all 2 4 1.33 0.67 0 ( -1 -1 -1 ) p42 cyclic-2dim-x 2 6 2.00 1.00 0 ( -1 -1 -1 ) p43 cyclic-2dim-all 2 6 2.00 1.00 0 ( -1 -1 -1 ) p44 cyclic-3dim-x 2 6 2.00 1.00 0 ( -1 -1 -1 ) p45 cyclic-3dim-all 2 6 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-1*3fix : 141.955 102.356 122.895 -> 141.955 -> 425.865 MByte/s p01 ring-1*3fix : 139.840 101.662 122.048 -> 139.840 -> 419.519 MByte/s p02 ring-1*3fix : 146.548 102.161 120.411 -> 146.548 -> 439.643 MByte/s p03 ring-1*3fix : 143.996 102.165 120.821 -> 143.996 -> 431.987 MByte/s p04 ring-1*3fix : 138.810 102.079 119.987 -> 138.810 -> 416.431 MByte/s p05 ring-1*3fix : 142.766 102.006 124.555 -> 142.766 -> 428.299 MByte/s p06 random-cyc-1dim : 140.822 101.996 122.876 -> 140.822 -> 422.467 MByte/s p07 random-cyc-1dim : 141.397 102.361 117.942 -> 141.397 -> 424.190 MByte/s p08 random-cyc-1dim : 139.663 102.140 123.212 -> 139.663 -> 418.988 MByte/s p09 random-cyc-1dim : 145.100 102.297 125.233 -> 145.100 -> 435.300 MByte/s p10 random-cyc-1dim : 141.623 102.052 124.008 -> 141.623 -> 424.869 MByte/s p11 random-cyc-1dim : 147.159 102.215 120.426 -> 147.159 -> 441.476 MByte/s p12 random-cyc-1dim : 148.353 102.252 120.427 -> 148.353 -> 445.060 MByte/s p13 random-cyc-1dim : 139.663 102.235 126.712 -> 139.663 -> 418.989 MByte/s p14 random-cyc-1dim : 142.126 102.139 118.130 -> 142.126 -> 426.379 MByte/s p15 random-cyc-1dim : 145.391 102.269 121.081 -> 145.391 -> 436.173 MByte/s p16 random-cyc-1dim : 140.318 102.145 121.850 -> 140.318 -> 420.955 MByte/s p17 random-cyc-1dim : 141.861 102.352 122.181 -> 141.861 -> 425.583 MByte/s p18 random-cyc-1dim : 143.309 102.101 123.117 -> 143.309 -> 429.928 MByte/s p19 random-cyc-1dim : 140.883 102.320 125.320 -> 140.883 -> 422.650 MByte/s p20 random-cyc-1dim : 148.721 102.590 119.615 -> 148.721 -> 446.163 MByte/s p21 random-cyc-1dim : 136.979 102.448 120.794 -> 136.979 -> 410.937 MByte/s p22 random-cyc-1dim : 144.727 102.390 118.614 -> 144.727 -> 434.180 MByte/s p23 random-cyc-1dim : 145.294 102.590 122.732 -> 145.294 -> 435.883 MByte/s p24 random-cyc-1dim : 144.258 102.584 126.521 -> 144.258 -> 432.775 MByte/s p25 random-cyc-1dim : 145.822 102.075 121.724 -> 145.822 -> 437.466 MByte/s p26 random-cyc-1dim : 141.669 102.218 121.659 -> 141.669 -> 425.008 MByte/s p27 random-cyc-1dim : 146.631 102.499 116.628 -> 146.631 -> 439.893 MByte/s p28 random-cyc-1dim : 138.869 102.286 125.243 -> 138.869 -> 416.606 MByte/s p29 random-cyc-1dim : 147.485 102.132 121.913 -> 147.485 -> 442.455 MByte/s p30 random-cyc-1dim : 148.362 102.163 124.488 -> 148.362 -> 445.086 MByte/s p31 random-cyc-1dim : 143.419 102.053 120.072 -> 143.419 -> 430.256 MByte/s p32 random-cyc-1dim : 141.336 102.150 120.389 -> 141.336 -> 424.009 MByte/s p33 random-cyc-1dim : 141.800 101.807 124.745 -> 141.800 -> 425.399 MByte/s p34 random-cyc-1dim : 145.040 101.894 119.131 -> 145.040 -> 435.119 MByte/s p35 random-cyc-1dim : 142.473 102.214 121.302 -> 142.473 -> 427.419 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 142.879 102.356 122.418 -> 142.879 -> 428.638 MByte/s p37 best bi-section : 69.334 91.517 96.405 -> 96.405 -> 289.214 MByte/s p38 worst bi-section : 69.422 91.419 96.869 -> 96.869 -> 290.607 MByte/s p39 one PingPong Pair : 69.413 0.000 0.000 -> 69.413 -> 208.239 MByte/s p40 acyclic-2dim-all : 102.886 98.477 93.200 -> 102.886 -> 308.658 MByte/s p41 acyclic-3dim-all : 102.637 98.403 92.930 -> 102.637 -> 307.912 MByte/s p42 cyclic-2dim-x : 144.223 102.103 122.924 -> 144.223 -> 432.668 MByte/s p43 cyclic-2dim-all : 141.062 102.050 123.680 -> 141.062 -> 423.187 MByte/s p44 cyclic-3dim-x : 146.236 102.346 122.369 -> 146.236 -> 438.709 MByte/s p45 cyclic-3dim-all : 143.181 102.317 123.934 -> 143.181 -> 429.542 MByte/s log_avg of all rings : 142.296 102.071 121.776 || 142.296 -> 426.888 MByte/s log_avg of all random : 143.320 102.232 121.909 || 143.320 -> 429.960 MByte/s log_avg(ring,random) : 142.807 102.152 121.843 ||(142.807 -> 428.421)MByte/s * size -> accumulated on all pr.: 428.421 306.455 365.528 ||(428.421)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-1*3fix : 128.179 136.620 134.913 -> 136.620 -> 409.861 MByte/s p01 ring-1*3fix : 129.006 134.376 136.397 -> 136.397 -> 409.191 MByte/s p02 ring-1*3fix : 134.327 142.554 138.808 -> 142.554 -> 427.662 MByte/s p03 ring-1*3fix : 132.754 138.067 135.291 -> 138.067 -> 414.200 MByte/s p04 ring-1*3fix : 136.410 133.540 134.850 -> 136.410 -> 409.230 MByte/s p05 ring-1*3fix : 137.663 138.555 131.912 -> 138.555 -> 415.666 MByte/s p06 random-cyc-1dim : 132.647 133.621 136.046 -> 136.046 -> 408.137 MByte/s p07 random-cyc-1dim : 132.119 130.931 136.972 -> 136.972 -> 410.916 MByte/s p08 random-cyc-1dim : 129.644 132.109 138.968 -> 138.968 -> 416.905 MByte/s p09 random-cyc-1dim : 131.653 141.868 137.790 -> 141.868 -> 425.604 MByte/s p10 random-cyc-1dim : 139.546 133.669 128.965 -> 139.546 -> 418.638 MByte/s p11 random-cyc-1dim : 132.154 142.630 140.035 -> 142.630 -> 427.890 MByte/s p12 random-cyc-1dim : 128.533 139.608 135.953 -> 139.608 -> 418.825 MByte/s p13 random-cyc-1dim : 133.544 134.554 132.761 -> 134.554 -> 403.662 MByte/s p14 random-cyc-1dim : 139.060 134.993 131.413 -> 139.060 -> 417.179 MByte/s p15 random-cyc-1dim : 144.399 138.574 133.264 -> 144.399 -> 433.198 MByte/s p16 random-cyc-1dim : 135.540 133.145 133.312 -> 135.540 -> 406.620 MByte/s p17 random-cyc-1dim : 133.878 133.165 135.420 -> 135.420 -> 406.261 MByte/s p18 random-cyc-1dim : 136.322 132.359 133.083 -> 136.322 -> 408.967 MByte/s p19 random-cyc-1dim : 132.500 135.911 132.077 -> 135.911 -> 407.732 MByte/s p20 random-cyc-1dim : 132.756 140.840 141.751 -> 141.751 -> 425.253 MByte/s p21 random-cyc-1dim : 133.961 135.242 132.667 -> 135.242 -> 405.725 MByte/s p22 random-cyc-1dim : 141.029 132.046 139.723 -> 141.029 -> 423.086 MByte/s p23 random-cyc-1dim : 139.626 134.471 135.472 -> 139.626 -> 418.878 MByte/s p24 random-cyc-1dim : 138.276 134.192 135.854 -> 138.276 -> 414.827 MByte/s p25 random-cyc-1dim : 132.496 133.245 136.107 -> 136.107 -> 408.320 MByte/s p26 random-cyc-1dim : 134.590 135.469 133.604 -> 135.469 -> 406.406 MByte/s p27 random-cyc-1dim : 134.372 138.833 142.131 -> 142.131 -> 426.394 MByte/s p28 random-cyc-1dim : 137.776 131.115 135.933 -> 137.776 -> 413.327 MByte/s p29 random-cyc-1dim : 133.870 136.025 145.233 -> 145.233 -> 435.698 MByte/s p30 random-cyc-1dim : 141.995 133.768 132.435 -> 141.995 -> 425.985 MByte/s p31 random-cyc-1dim : 140.226 134.055 133.447 -> 140.226 -> 420.678 MByte/s p32 random-cyc-1dim : 132.332 136.551 138.196 -> 138.196 -> 414.588 MByte/s p33 random-cyc-1dim : 132.114 131.946 141.411 -> 141.411 -> 424.233 MByte/s p34 random-cyc-1dim : 140.622 134.168 129.552 -> 140.622 -> 421.865 MByte/s p35 random-cyc-1dim : 135.438 135.094 136.879 -> 136.879 -> 410.636 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim : 133.227 132.347 134.966 -> 134.966 -> 404.899 MByte/s p37 best bi-section : 93.696 95.793 94.508 -> 95.793 -> 287.380 MByte/s p38 worst bi-section : 93.365 96.444 95.146 -> 96.444 -> 289.333 MByte/s p39 one PingPong Pair : 69.139 69.071 68.810 -> 69.139 -> 207.417 MByte/s p40 acyclic-2dim-all : 102.692 101.287 101.345 -> 102.692 -> 308.075 MByte/s p41 acyclic-3dim-all : 101.578 102.150 102.357 -> 102.357 -> 307.072 MByte/s p42 cyclic-2dim-x : 138.361 137.324 139.722 -> 139.722 -> 419.167 MByte/s p43 cyclic-2dim-all : 130.516 132.274 137.326 -> 137.326 -> 411.978 MByte/s p44 cyclic-3dim-x : 137.580 139.773 143.521 -> 143.521 -> 430.563 MByte/s p45 cyclic-3dim-all : 134.585 131.575 132.316 -> 134.585 -> 403.756 MByte/s log_avg of all rings : 133.010 137.253 135.346 || 138.084 -> 414.252 MByte/s log_avg of all random : 135.379 135.106 135.829 || 138.931 -> 416.793 MByte/s log_avg(ring,random) : 134.189 136.176 135.588 ||(138.507 -> 415.520)MByte/s * size -> accumulated on all pr.: 402.568 408.527 406.763 ||(415.520)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-1*3fix p00 method 0 =Sndrcv :( 26.487) 0.038 0.597 8.791 66.268 343.587 514.875 -> 141.955 -> 425.865 MByte/s p00 method 1 =Alltoal :( 36.020) 0.028 0.415 6.129 53.167 289.670 335.793 -> 102.356 -> 307.069 MByte/s p00 method 2 =non-blk :( 46.100) 0.022 0.337 5.106 50.706 275.780 429.084 -> 122.895 -> 368.685 MByte/s p01 ring-1*3fix p01 method 0 =Sndrcv :( 26.044) 0.038 0.598 8.777 66.268 423.086 449.767 -> 139.840 -> 419.519 MByte/s p01 method 1 =Alltoal :( 36.272) 0.028 0.413 6.111 53.154 285.910 329.436 -> 101.662 -> 304.987 MByte/s p01 method 2 =non-blk :( 45.767) 0.022 0.336 5.110 51.384 319.085 433.049 -> 122.048 -> 366.145 MByte/s p02 ring-1*3fix p02 method 0 =Sndrcv :( 26.287) 0.038 0.598 8.780 66.416 429.944 475.680 -> 146.548 -> 439.643 MByte/s p02 method 1 =Alltoal :( 36.183) 0.028 0.413 6.135 53.169 287.018 331.618 -> 102.161 -> 306.482 MByte/s p02 method 2 =non-blk :( 46.151) 0.022 0.336 5.107 51.237 279.100 433.722 -> 120.411 -> 361.232 MByte/s p03 ring-1*3fix p03 method 0 =Sndrcv :( 26.044) 0.038 0.597 8.794 66.268 341.915 450.033 -> 143.996 -> 431.987 MByte/s p03 method 1 =Alltoal :( 36.481) 0.027 0.415 6.093 53.168 286.055 331.205 -> 102.165 -> 306.495 MByte/s p03 method 2 =non-blk :( 45.760) 0.022 0.337 5.107 50.657 272.078 434.362 -> 120.821 -> 362.464 MByte/s p04 ring-1*3fix p04 method 0 =Sndrcv :( 26.277) 0.038 0.599 8.797 66.378 373.056 448.805 -> 138.810 -> 416.431 MByte/s p04 method 1 =Alltoal :( 36.154) 0.028 0.415 6.062 53.195 287.833 331.671 -> 102.079 -> 306.237 MByte/s p04 method 2 =non-blk :( 46.179) 0.022 0.333 5.105 51.383 285.240 435.445 -> 119.987 -> 359.961 MByte/s p05 ring-1*3fix p05 method 0 =Sndrcv :( 26.062) 0.038 0.598 8.780 66.340 342.952 449.804 -> 142.766 -> 428.299 MByte/s p05 method 1 =Alltoal :( 36.273) 0.028 0.416 6.050 53.182 288.576 331.435 -> 102.006 -> 306.019 MByte/s p05 method 2 =non-blk :( 45.897) 0.022 0.334 5.106 51.106 289.859 439.183 -> 124.555 -> 373.666 MByte/s p06 random-cyc-1dim p06 method 0 =Sndrcv :( 26.086) 0.038 0.599 8.780 66.268 429.573 451.134 -> 140.822 -> 422.467 MByte/s p06 method 1 =Alltoal :( 36.029) 0.028 0.417 6.065 53.195 286.020 330.945 -> 101.996 -> 305.988 MByte/s p06 method 2 =non-blk :( 46.133) 0.022 0.332 5.105 51.430 268.902 454.150 -> 122.876 -> 368.629 MByte/s p07 random-cyc-1dim p07 method 0 =Sndrcv :( 26.262) 0.038 0.599 8.780 66.305 350.895 450.045 -> 141.397 -> 424.190 MByte/s p07 method 1 =Alltoal :( 36.218) 0.028 0.415 6.131 53.070 288.020 330.293 -> 102.361 -> 307.083 MByte/s p07 method 2 =non-blk :( 45.801) 0.022 0.332 5.096 50.522 272.623 433.196 -> 117.942 -> 353.825 MByte/s p08 random-cyc-1dim p08 method 0 =Sndrcv :( 26.074) 0.038 0.599 8.787 66.323 300.734 446.987 -> 139.663 -> 418.988 MByte/s p08 method 1 =Alltoal :( 36.125) 0.028 0.417 6.043 53.014 286.904 328.161 -> 102.140 -> 306.421 MByte/s p08 method 2 =non-blk :( 46.109) 0.022 0.333 5.101 50.389 289.762 443.103 -> 123.212 -> 369.636 MByte/s p09 random-cyc-1dim p09 method 0 =Sndrcv :( 26.247) 0.038 0.599 8.794 66.175 430.822 449.465 -> 145.100 -> 435.300 MByte/s p09 method 1 =Alltoal :( 36.068) 0.028 0.415 6.073 52.982 286.979 329.670 -> 102.297 -> 306.892 MByte/s p09 method 2 =non-blk :( 45.836) 0.022 0.333 5.096 51.458 288.441 444.771 -> 125.233 -> 375.700 MByte/s p10 random-cyc-1dim p10 method 0 =Sndrcv :( 26.105) 0.038 0.597 8.784 66.358 339.573 449.876 -> 141.623 -> 424.869 MByte/s p10 method 1 =Alltoal :( 36.029) 0.028 0.418 6.082 52.988 286.829 330.787 -> 102.052 -> 306.155 MByte/s p10 method 2 =non-blk :( 45.861) 0.022 0.335 5.106 51.078 288.056 435.353 -> 124.008 -> 372.025 MByte/s p11 random-cyc-1dim p11 method 0 =Sndrcv :( 26.290) 0.038 0.596 8.675 66.138 430.584 447.583 -> 147.159 -> 441.476 MByte/s p11 method 1 =Alltoal :( 36.116) 0.028 0.415 6.075 53.195 287.909 331.055 -> 102.215 -> 306.644 MByte/s p11 method 2 =non-blk :( 45.904) 0.022 0.336 5.098 51.319 308.539 435.104 -> 120.426 -> 361.277 MByte/s p12 random-cyc-1dim p12 method 0 =Sndrcv :( 26.049) 0.038 0.598 8.678 66.305 395.346 474.967 -> 148.353 -> 445.060 MByte/s p12 method 1 =Alltoal :( 36.039) 0.028 0.413 6.139 53.023 287.869 330.443 -> 102.252 -> 306.756 MByte/s p12 method 2 =non-blk :( 45.840) 0.022 0.336 5.109 51.476 267.257 431.157 -> 120.427 -> 361.282 MByte/s p13 random-cyc-1dim p13 method 0 =Sndrcv :( 26.267) 0.038 0.598 8.684 66.396 308.621 450.069 -> 139.663 -> 418.989 MByte/s p13 method 1 =Alltoal :( 36.250) 0.028 0.415 6.067 53.154 288.282 331.167 -> 102.235 -> 306.704 MByte/s p13 method 2 =non-blk :( 45.898) 0.022 0.337 5.099 51.049 304.908 431.680 -> 126.712 -> 380.136 MByte/s p14 random-cyc-1dim p14 method 0 =Sndrcv :( 26.019) 0.038 0.598 8.649 66.193 373.247 451.948 -> 142.126 -> 426.379 MByte/s p14 method 1 =Alltoal :( 35.991) 0.028 0.416 6.080 53.195 287.607 331.212 -> 102.139 -> 306.417 MByte/s p14 method 2 =non-blk :( 45.580) 0.022 0.336 5.085 51.410 280.037 433.340 -> 118.130 -> 354.391 MByte/s p15 random-cyc-1dim p15 method 0 =Sndrcv :( 26.259) 0.038 0.599 8.536 66.398 426.860 451.170 -> 145.391 -> 436.173 MByte/s p15 method 1 =Alltoal :( 36.057) 0.028 0.416 6.139 53.195 287.682 330.052 -> 102.269 -> 306.806 MByte/s p15 method 2 =non-blk :( 45.788) 0.022 0.337 5.076 50.458 267.353 437.979 -> 121.081 -> 363.242 MByte/s p16 random-cyc-1dim p16 method 0 =Sndrcv :( 26.024) 0.038 0.597 8.502 66.213 368.514 450.601 -> 140.318 -> 420.955 MByte/s p16 method 1 =Alltoal :( 35.961) 0.028 0.414 6.108 53.126 282.029 331.533 -> 102.145 -> 306.435 MByte/s p16 method 2 =non-blk :( 45.560) 0.022 0.336 5.085 51.008 265.537 434.226 -> 121.850 -> 365.549 MByte/s p17 random-cyc-1dim p17 method 0 =Sndrcv :( 26.148) 0.038 0.599 8.505 66.285 429.202 450.286 -> 141.861 -> 425.583 MByte/s p17 method 1 =Alltoal :( 36.096) 0.028 0.414 6.131 52.971 287.124 330.787 -> 102.352 -> 307.057 MByte/s p17 method 2 =non-blk :( 45.829) 0.022 0.337 5.104 51.466 273.167 437.055 -> 122.181 -> 366.544 MByte/s p18 random-cyc-1dim p18 method 0 =Sndrcv :( 26.012) 0.038 0.599 8.511 66.378 354.767 474.496 -> 143.309 -> 429.928 MByte/s p18 method 1 =Alltoal :( 36.029) 0.028 0.416 6.133 52.975 286.129 329.333 -> 102.101 -> 306.303 MByte/s p18 method 2 =non-blk :( 45.526) 0.022 0.331 5.092 51.068 282.784 433.687 -> 123.117 -> 369.351 MByte/s p19 random-cyc-1dim p19 method 0 =Sndrcv :( 26.142) 0.038 0.598 8.665 66.268 320.839 449.166 -> 140.883 -> 422.650 MByte/s p19 method 1 =Alltoal :( 36.106) 0.028 0.414 6.128 53.207 286.758 331.959 -> 102.320 -> 306.960 MByte/s p19 method 2 =non-blk :( 46.067) 0.022 0.332 5.098 51.374 311.010 442.368 -> 125.320 -> 375.961 MByte/s p20 random-cyc-1dim p20 method 0 =Sndrcv :( 26.024) 0.038 0.600 8.662 66.416 428.832 476.206 -> 148.721 -> 446.163 MByte/s p20 method 1 =Alltoal :( 36.202) 0.028 0.414 6.158 53.182 288.353 330.599 -> 102.590 -> 307.769 MByte/s p20 method 2 =non-blk :( 45.553) 0.022 0.331 5.045 51.458 277.669 435.071 -> 119.615 -> 358.844 MByte/s p21 random-cyc-1dim p21 method 0 =Sndrcv :( 26.217) 0.038 0.599 8.668 66.380 343.739 449.778 -> 136.979 -> 410.937 MByte/s p21 method 1 =Alltoal :( 35.991) 0.028 0.413 6.144 53.126 285.945 330.697 -> 102.448 -> 307.343 MByte/s p21 method 2 =non-blk :( 46.130) 0.022 0.332 5.034 51.068 270.351 437.248 -> 120.794 -> 362.383 MByte/s p22 random-cyc-1dim p22 method 0 =Sndrcv :( 26.038) 0.038 0.598 8.781 66.248 427.594 448.433 -> 144.727 -> 434.180 MByte/s p22 method 1 =Alltoal :( 35.971) 0.028 0.413 6.124 53.114 287.718 330.559 -> 102.390 -> 307.169 MByte/s p22 method 2 =non-blk :( 45.787) 0.022 0.333 4.960 51.458 270.718 434.316 -> 118.614 -> 355.843 MByte/s p23 random-cyc-1dim p23 method 0 =Sndrcv :( 26.247) 0.038 0.599 8.794 66.287 430.331 448.026 -> 145.294 -> 435.883 MByte/s p23 method 1 =Alltoal :( 36.038) 0.028 0.414 6.138 53.152 288.091 330.136 -> 102.590 -> 307.771 MByte/s p23 method 2 =non-blk :( 46.123) 0.022 0.336 4.972 51.040 288.247 429.512 -> 122.732 -> 368.195 MByte/s p24 random-cyc-1dim p24 method 0 =Sndrcv :( 26.037) 0.038 0.600 8.811 66.155 361.594 488.989 -> 144.258 -> 432.775 MByte/s p24 method 1 =Alltoal :( 36.154) 0.028 0.413 6.135 53.167 289.634 330.866 -> 102.584 -> 307.752 MByte/s p24 method 2 =non-blk :( 45.773) 0.022 0.336 4.981 51.374 281.708 436.520 -> 126.521 -> 379.562 MByte/s p25 random-cyc-1dim p25 method 0 =Sndrcv :( 26.235) 0.038 0.598 8.798 66.248 427.095 450.395 -> 145.822 -> 437.466 MByte/s p25 method 1 =Alltoal :( 36.144) 0.028 0.413 6.160 53.207 287.461 329.514 -> 102.075 -> 306.226 MByte/s p25 method 2 =non-blk :( 46.137) 0.022 0.337 4.970 51.402 325.768 443.888 -> 121.724 -> 365.172 MByte/s p26 random-cyc-1dim p26 method 0 =Sndrcv :( 26.074) 0.038 0.599 8.798 66.398 395.761 451.122 -> 141.669 -> 425.008 MByte/s p26 method 1 =Alltoal :( 35.981) 0.028 0.413 6.104 52.933 286.864 331.277 -> 102.218 -> 306.653 MByte/s p26 method 2 =non-blk :( 45.861) 0.022 0.337 5.000 51.050 267.096 431.126 -> 121.659 -> 364.978 MByte/s p27 random-cyc-1dim p27 method 0 =Sndrcv :( 26.241) 0.038 0.599 8.777 66.305 429.336 449.068 -> 146.631 -> 439.893 MByte/s p27 method 1 =Alltoal :( 35.932) 0.028 0.414 6.146 52.969 287.160 330.899 -> 102.499 -> 307.496 MByte/s p27 method 2 =non-blk :( 46.041) 0.022 0.336 4.986 50.909 269.068 432.014 -> 116.628 -> 349.883 MByte/s p28 random-cyc-1dim p28 method 0 =Sndrcv :( 26.056) 0.038 0.598 8.797 66.285 369.620 449.226 -> 138.869 -> 416.606 MByte/s p28 method 1 =Alltoal :( 36.018) 0.028 0.414 6.139 52.930 286.020 330.039 -> 102.286 -> 306.857 MByte/s p28 method 2 =non-blk :( 46.060) 0.022 0.336 4.985 51.458 288.242 426.543 -> 125.243 -> 375.728 MByte/s p29 random-cyc-1dim p29 method 0 =Sndrcv :( 26.037) 0.038 0.599 8.791 66.340 426.978 449.490 -> 147.485 -> 442.455 MByte/s p29 method 1 =Alltoal :( 36.057) 0.028 0.417 6.117 53.154 285.215 330.925 -> 102.132 -> 306.396 MByte/s p29 method 2 =non-blk :( 46.073) 0.022 0.336 4.983 51.383 276.839 438.701 -> 121.913 -> 365.740 MByte/s p30 random-cyc-1dim p30 method 0 =Sndrcv :( 26.049) 0.038 0.599 8.787 66.230 431.076 449.224 -> 148.362 -> 445.086 MByte/s p30 method 1 =Alltoal :( 36.298) 0.028 0.417 6.133 53.010 287.199 331.368 -> 102.163 -> 306.488 MByte/s p30 method 2 =non-blk :( 46.068) 0.022 0.337 5.020 51.439 272.767 431.291 -> 124.488 -> 373.464 MByte/s p31 random-cyc-1dim p31 method 0 =Sndrcv :( 26.030) 0.038 0.599 8.777 66.213 430.078 488.207 -> 143.419 -> 430.256 MByte/s p31 method 1 =Alltoal :( 36.225) 0.028 0.415 6.032 53.113 286.164 331.009 -> 102.053 -> 306.158 MByte/s p31 method 2 =non-blk :( 45.920) 0.022 0.332 5.038 51.069 293.098 443.842 -> 120.072 -> 360.215 MByte/s p32 random-cyc-1dim p32 method 0 =Sndrcv :( 26.037) 0.038 0.600 8.787 66.323 430.450 467.709 -> 141.336 -> 424.009 MByte/s p32 method 1 =Alltoal :( 36.029) 0.028 0.416 6.026 53.001 285.468 329.314 -> 102.150 -> 306.451 MByte/s p32 method 2 =non-blk :( 45.994) 0.022 0.330 5.067 50.891 278.354 443.912 -> 120.389 -> 361.168 MByte/s p33 random-cyc-1dim p33 method 0 =Sndrcv :( 26.049) 0.038 0.599 8.767 66.268 426.378 450.650 -> 141.800 -> 425.399 MByte/s p33 method 1 =Alltoal :( 36.077) 0.028 0.414 6.015 53.075 270.936 330.052 -> 101.807 -> 305.420 MByte/s p33 method 2 =non-blk :( 45.847) 0.022 0.332 5.085 51.475 286.683 428.755 -> 124.745 -> 374.236 MByte/s p34 random-cyc-1dim p34 method 0 =Sndrcv :( 26.068) 0.038 0.600 8.777 66.287 395.233 451.498 -> 145.040 -> 435.119 MByte/s p34 method 1 =Alltoal :( 36.096) 0.028 0.415 6.030 52.961 287.500 330.775 -> 101.894 -> 305.683 MByte/s p34 method 2 =non-blk :( 45.939) 0.022 0.332 5.084 51.355 270.661 438.172 -> 119.131 -> 357.393 MByte/s p35 random-cyc-1dim p35 method 0 =Sndrcv :( 25.982) 0.038 0.599 8.777 66.195 367.426 445.775 -> 142.473 -> 427.419 MByte/s p35 method 1 =Alltoal :( 36.116) 0.028 0.414 5.916 52.996 289.782 330.092 -> 102.214 -> 306.641 MByte/s p35 method 2 =non-blk :( 45.861) 0.022 0.332 5.117 50.657 275.529 438.782 -> 121.302 -> 363.905 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim p36 method 0 =Sndrcv :( 26.061) 0.038 0.600 8.777 66.230 368.700 447.596 -> 142.879 -> 428.638 MByte/s p36 method 1 =Alltoal :( 36.039) 0.028 0.413 5.949 52.422 287.904 330.794 -> 102.356 -> 307.067 MByte/s p36 method 2 =non-blk :( 45.918) 0.022 0.336 5.105 51.063 284.268 433.654 -> 122.418 -> 367.254 MByte/s p37 best bi-section p37 method 0 =Sndrcv :( 22.932) 0.015 0.215 3.161 27.028 164.271 265.282 -> 69.334 -> 208.003 MByte/s p37 method 1 =Alltoal :( 35.388) 0.009 0.148 2.245 25.732 257.271 338.011 -> 91.517 -> 274.552 MByte/s p37 method 2 =non-blk :( 24.911) 0.013 0.205 3.232 33.242 269.523 337.991 -> 96.405 -> 289.214 MByte/s p38 worst bi-section p38 method 0 =Sndrcv :( 22.932) 0.015 0.216 3.176 27.270 163.732 265.294 -> 69.422 -> 208.267 MByte/s p38 method 1 =Alltoal :( 35.172) 0.009 0.148 2.290 25.761 256.032 337.422 -> 91.419 -> 274.257 MByte/s p38 method 2 =non-blk :( 24.900) 0.013 0.202 3.175 33.185 275.146 338.442 -> 96.869 -> 290.607 MByte/s p39 one PingPong Pair p39 method 0 =Sndrcv :( 22.883) 0.015 0.216 3.171 27.279 164.656 265.710 -> 69.413 -> 208.239 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 :( 26.062) 0.026 0.397 5.850 44.514 288.313 346.978 -> 102.886 -> 308.658 MByte/s p40 method 1 =Alltoal :( 36.009) 0.019 0.282 4.244 39.010 278.271 333.477 -> 98.477 -> 295.431 MByte/s p40 method 2 =non-blk :( 43.472) 0.015 0.236 3.613 35.788 233.125 328.405 -> 93.200 -> 279.601 MByte/s p41 acyclic-3dim-all p41 method 0 =Sndrcv :( 26.037) 0.026 0.397 5.865 44.513 288.223 345.307 -> 102.637 -> 307.912 MByte/s p41 method 1 =Alltoal :( 35.866) 0.019 0.283 4.255 39.009 277.750 333.071 -> 98.403 -> 295.209 MByte/s p41 method 2 =non-blk :( 43.387) 0.015 0.237 3.571 36.052 234.935 330.052 -> 92.930 -> 278.791 MByte/s p42 cyclic-2dim-x p42 method 0 =Sndrcv :( 26.049) 0.038 0.598 8.790 66.305 429.825 450.214 -> 144.223 -> 432.668 MByte/s p42 method 1 =Alltoal :( 36.188) 0.028 0.413 6.039 52.419 288.353 331.153 -> 102.103 -> 306.310 MByte/s p42 method 2 =non-blk :( 45.850) 0.022 0.336 5.102 51.355 267.161 446.689 -> 122.924 -> 368.773 MByte/s p43 cyclic-2dim-all p43 method 0 =Sndrcv :( 26.019) 0.038 0.598 8.784 65.338 347.047 445.847 -> 141.062 -> 423.187 MByte/s p43 method 1 =Alltoal :( 35.981) 0.028 0.414 6.067 53.195 287.014 327.533 -> 102.050 -> 306.150 MByte/s p43 method 2 =non-blk :( 45.707) 0.022 0.337 5.097 50.585 276.488 430.494 -> 123.680 -> 371.040 MByte/s p44 cyclic-3dim-x p44 method 0 =Sndrcv :( 26.074) 0.038 0.599 8.797 66.303 430.703 451.668 -> 146.236 -> 438.709 MByte/s p44 method 1 =Alltoal :( 35.932) 0.028 0.412 6.082 53.181 286.314 331.042 -> 102.346 -> 307.039 MByte/s p44 method 2 =non-blk :( 45.816) 0.022 0.336 5.109 51.623 288.916 430.152 -> 122.369 -> 367.107 MByte/s p45 cyclic-3dim-all p45 method 0 =Sndrcv :( 25.988) 0.038 0.598 8.794 66.360 312.058 451.474 -> 143.181 -> 429.542 MByte/s p45 method 1 =Alltoal :( 35.981) 0.028 0.415 5.973 53.182 287.385 330.840 -> 102.317 -> 306.951 MByte/s p45 method 2 =non-blk :( 45.813) 0.022 0.336 5.098 51.328 278.431 432.146 -> 123.934 -> 371.803 MByte/s log_avg of all rings - ring, method 0 = Sndrcv : 0.038 0.598 8.787 66.323 373.939 464.218 || 142.296 -> 426.888 MByte/s - ring, method 1 = Alltoal: 0.028 0.414 6.096 53.172 287.507 331.854 || 102.071 -> 306.214 MByte/s - ring, method 2 = non-blk: 0.022 0.336 5.107 51.078 286.453 434.130 || 121.776 -> 365.328 MByte/s log_avg of all random - random, method 0 = Sndrcv : 0.038 0.599 8.722 66.286 390.363 455.231 || 143.320 -> 429.960 MByte/s - random, method 1 = Alltoal: 0.028 0.415 6.097 53.079 286.503 330.516 || 102.232 -> 306.696 MByte/s - random, method 2 = non-blk: 0.022 0.334 5.056 51.165 281.678 436.523 || 121.909 -> 365.728 MByte/s log_avg(ring,random) - average, method 0 = Sndrcv : 0.038 0.598 8.754 66.304 382.062 459.702 || 142.807 -> 428.421 MByte/s - average, method 1 = Alltoal: 0.028 0.415 6.097 53.126 287.005 331.185 || 102.152 -> 306.455 MByte/s - average, method 2 = non-blk: 0.022 0.335 5.081 51.122 284.055 435.325 || 121.843 -> 365.528 MByte/s * size -> accumulated on all processes: - accumulated, mthd 0 = Sndrcv : 0.115 1.795 26.263 198.913 1146.187 1379.106 || 428.421 MByte/s - accumulated, mthd 1 = Alltoal: 0.083 1.244 18.290 159.377 861.014 993.554 || 306.455 MByte/s - accumulated, mthd 2 = non-blk: 0.065 1.005 15.243 153.365 852.166 1305.976 || 365.528 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 0.115 0.038 0.038 0.038 0.038 0.028 0.022 2 0.229 0.076 0.076 0.076 0.076 0.055 0.044 4 0.456 0.152 0.152 0.152 0.152 0.110 0.087 8 0.919 0.306 0.307 0.306 0.306 0.221 0.175 16 1.795 0.598 0.598 0.599 0.598 0.415 0.335 32 3.547 1.182 1.182 1.182 1.182 0.819 0.665 64 6.946 2.315 2.317 2.314 2.315 1.595 1.317 128 13.226 4.409 4.367 4.450 4.409 3.061 2.580 256 26.263 8.754 8.787 8.722 8.754 6.097 5.081 512 51.259 17.086 17.006 17.167 17.086 11.937 9.991 1024 100.530 33.510 33.401 33.619 33.510 23.384 19.729 2048 122.073 40.691 40.703 40.678 40.691 31.232 29.950 4096 198.913 66.304 66.323 66.286 66.304 53.126 51.122 10624 363.537 121.179 121.016 121.342 121.179 81.465 101.692 27554 569.808 189.936 185.781 194.184 189.141 134.679 167.401 71468 855.060 285.020 285.375 284.666 285.020 197.405 205.639 185364 1146.187 382.062 373.939 390.363 382.062 287.005 284.055 480774 1208.865 402.955 399.219 406.726 402.955 308.292 334.331 1246974 1446.418 482.139 478.011 486.304 477.988 334.277 453.290 3234251 1498.350 499.450 499.304 499.596 499.402 338.742 453.591 8388608 1379.260 459.753 464.218 455.332 459.702 331.185 435.325 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-1*3fix :( 26.487) 0.038 0.597 8.791 66.268 343.587 514.875 -> 141.955 -> 425.865 MByte/s p01 ring-1*3fix :( 26.044) 0.038 0.598 8.777 66.268 423.086 449.767 -> 139.840 -> 419.519 MByte/s p02 ring-1*3fix :( 26.287) 0.038 0.598 8.780 66.416 429.944 475.680 -> 146.548 -> 439.643 MByte/s p03 ring-1*3fix :( 26.044) 0.038 0.597 8.794 66.268 341.915 450.033 -> 143.996 -> 431.987 MByte/s p04 ring-1*3fix :( 26.277) 0.038 0.599 8.797 66.378 373.056 448.805 -> 138.887 -> 416.660 MByte/s p05 ring-1*3fix :( 26.062) 0.038 0.598 8.780 66.340 342.952 449.804 -> 142.766 -> 428.299 MByte/s p06 random-cyc-1dim :( 26.086) 0.038 0.599 8.780 66.268 429.573 454.150 -> 141.999 -> 425.998 MByte/s p07 random-cyc-1dim :( 26.262) 0.038 0.599 8.780 66.305 350.895 450.045 -> 141.397 -> 424.190 MByte/s p08 random-cyc-1dim :( 26.074) 0.038 0.599 8.787 66.323 300.734 446.987 -> 139.663 -> 418.988 MByte/s p09 random-cyc-1dim :( 26.247) 0.038 0.599 8.794 66.175 430.822 449.465 -> 147.367 -> 442.101 MByte/s p10 random-cyc-1dim :( 26.105) 0.038 0.597 8.784 66.358 339.573 449.876 -> 141.623 -> 424.869 MByte/s p11 random-cyc-1dim :( 26.290) 0.038 0.596 8.675 66.138 430.584 447.583 -> 147.159 -> 441.476 MByte/s p12 random-cyc-1dim :( 26.049) 0.038 0.598 8.678 66.305 395.346 474.967 -> 148.353 -> 445.060 MByte/s p13 random-cyc-1dim :( 26.267) 0.038 0.598 8.684 66.396 308.621 450.069 -> 142.127 -> 426.382 MByte/s p14 random-cyc-1dim :( 26.019) 0.038 0.598 8.649 66.193 373.247 451.948 -> 142.126 -> 426.379 MByte/s p15 random-cyc-1dim :( 26.259) 0.038 0.599 8.536 66.398 426.860 451.170 -> 145.518 -> 436.555 MByte/s p16 random-cyc-1dim :( 26.024) 0.038 0.597 8.502 66.213 368.514 450.601 -> 142.201 -> 426.603 MByte/s p17 random-cyc-1dim :( 26.148) 0.038 0.599 8.505 66.285 429.202 450.286 -> 141.861 -> 425.583 MByte/s p18 random-cyc-1dim :( 26.012) 0.038 0.599 8.511 66.378 354.767 474.496 -> 145.774 -> 437.322 MByte/s p19 random-cyc-1dim :( 26.142) 0.038 0.598 8.665 66.268 320.839 449.166 -> 140.898 -> 422.693 MByte/s p20 random-cyc-1dim :( 26.024) 0.038 0.600 8.662 66.416 428.832 476.206 -> 148.721 -> 446.163 MByte/s p21 random-cyc-1dim :( 26.217) 0.038 0.599 8.668 66.380 343.739 449.778 -> 137.073 -> 411.218 MByte/s p22 random-cyc-1dim :( 26.038) 0.038 0.598 8.781 66.248 427.594 448.433 -> 144.727 -> 434.180 MByte/s p23 random-cyc-1dim :( 26.247) 0.038 0.599 8.794 66.287 430.331 448.026 -> 145.740 -> 437.219 MByte/s p24 random-cyc-1dim :( 26.037) 0.038 0.600 8.811 66.155 361.594 488.989 -> 144.395 -> 433.184 MByte/s p25 random-cyc-1dim :( 26.235) 0.038 0.598 8.798 66.248 427.095 450.395 -> 145.822 -> 437.466 MByte/s p26 random-cyc-1dim :( 26.074) 0.038 0.599 8.798 66.398 395.761 451.122 -> 141.669 -> 425.008 MByte/s p27 random-cyc-1dim :( 26.241) 0.038 0.599 8.777 66.305 429.336 449.068 -> 146.631 -> 439.893 MByte/s p28 random-cyc-1dim :( 26.056) 0.038 0.598 8.797 66.285 369.620 449.226 -> 140.948 -> 422.845 MByte/s p29 random-cyc-1dim :( 26.037) 0.038 0.599 8.791 66.340 426.978 449.490 -> 147.485 -> 442.455 MByte/s p30 random-cyc-1dim :( 26.049) 0.038 0.599 8.787 66.230 431.076 449.224 -> 148.362 -> 445.086 MByte/s p31 random-cyc-1dim :( 26.030) 0.038 0.599 8.777 66.213 430.078 488.207 -> 143.544 -> 430.633 MByte/s p32 random-cyc-1dim :( 26.037) 0.038 0.600 8.787 66.323 430.450 467.709 -> 141.365 -> 424.094 MByte/s p33 random-cyc-1dim :( 26.049) 0.038 0.599 8.767 66.268 426.378 450.650 -> 141.800 -> 425.399 MByte/s p34 random-cyc-1dim :( 26.068) 0.038 0.600 8.777 66.287 395.233 451.498 -> 145.040 -> 435.119 MByte/s p35 random-cyc-1dim :( 25.982) 0.038 0.599 8.777 66.195 367.426 445.775 -> 142.473 -> 427.419 MByte/s -- additional patterns that are not used to compute b_eff: p36 worst-cyc-1dim :( 26.061) 0.038 0.600 8.777 66.230 368.700 447.596 -> 142.879 -> 428.638 MByte/s p37 best bi-section :( 22.932) 0.015 0.215 3.232 33.242 269.523 338.011 -> 96.409 -> 289.227 MByte/s p38 worst bi-section :( 22.932) 0.015 0.216 3.176 33.185 275.146 338.442 -> 96.873 -> 290.619 MByte/s p39 one PingPong Pair :( 22.883) 0.015 0.216 3.171 27.279 164.656 265.710 -> 69.413 -> 208.239 MByte/s p40 acyclic-2dim-all :( 26.062) 0.026 0.397 5.850 44.514 288.313 346.978 -> 102.886 -> 308.658 MByte/s p41 acyclic-3dim-all :( 26.037) 0.026 0.397 5.865 44.513 288.223 345.307 -> 102.646 -> 307.937 MByte/s p42 cyclic-2dim-x :( 26.049) 0.038 0.598 8.790 66.305 429.825 450.214 -> 144.223 -> 432.668 MByte/s p43 cyclic-2dim-all :( 26.019) 0.038 0.598 8.784 65.338 347.047 445.847 -> 141.062 -> 423.187 MByte/s p44 cyclic-3dim-x :( 26.074) 0.038 0.599 8.797 66.303 430.703 451.668 -> 146.236 -> 438.709 MByte/s p45 cyclic-3dim-all :( 25.988) 0.038 0.598 8.794 66.360 312.058 451.474 -> 143.181 -> 429.542 MByte/s log_avg of all rings : 0.038 0.598 8.787 66.323 373.939 464.218 || 142.309 -> 426.927 MByte/s log_avg of all random : 0.038 0.599 8.722 66.286 390.363 455.332 || 143.766 -> 431.299 MByte/s log_avg(ring,random) : 0.038 0.598 8.754 66.304 382.062 459.753 || 143.036 -> 429.108 MByte/s * size -> accumulated on all pr.: 0.115 1.795 26.263 198.913 1146.187 1379.260 || 429.108 MByte/s SECTION-BY-PATTERN-MSGLNG-END SECTION-BEFF-BEGIN The effective bandwidth is b_eff = 429.108 MByte/s on 3 processes ( = 143.036 MByte/s * 3 processes) Ping-pong latency: 22.883 microsec Ping-pong bandwidth: 797.131 MByte/s at Lmax= 8.000 MByte (MByte/s=1e6 Byte/s) (MByte=2**20 Byte) system parameters : 3 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 13:42:18 1999 Total execution wall clock time = 65 seconds SECTION-BEFF-END b_eff = 429.108 MB/s = 143.036 * 3 PEs with 1024 MB/PE on HI-UX/MPP himiko 03-00 ad2b0 SR8000