Results for MPI perforance tests on sp2
Contents
memcpy
Determining delivered memory performance
Target "memcpy" is up to date.
Size (bytes) Time (sec) Rate (MB/sec)
4 0.000000 18.985667
8 0.000000 37.941215
16 0.000000 76.033340
32 0.000000 70.971692
64 0.000001 109.089921
128 0.000001 149.138266
256 0.000001 182.667099
512 0.000002 205.798686
1024 0.000005 219.715286
2048 0.000009 227.396829
4096 0.000018 231.439401
8192 0.000035 233.504672
16384 0.000070 234.559777
32768 0.000139 235.089146
65536 0.000278 235.349496
131072 0.000575 227.875352
262144 0.001592 164.673647
524288 0.003165 165.663595
1048576 0.006350 165.134634
2097152 0.012861 163.065122
Determining delivered memory performance with unaligned data
Target "memcpy" is up to date.
Size (bytes) Time (sec) Rate (MB/sec)
4 0.000000 15.642919
8 0.000000 29.534408
16 0.000000 56.050473
32 0.000001 56.050473
64 0.000001 86.870124
128 0.000001 119.810698
256 0.000002 147.840648
512 0.000003 167.422557
1024 0.000006 179.298277
2048 0.000011 185.893654
4096 0.000022 189.360754
8192 0.000043 191.146002
16384 0.000085 192.051594
32768 0.000170 192.514659
65536 0.000340 192.738759
131072 0.000698 187.766546
262144 0.001854 141.409004
524288 0.003684 142.316805
1048576 0.007365 142.375744
2097152 0.014940 140.367628
pingpong
Benchmarking point to point performance
Target "pingpong" is up to date.
Kind n time (sec) Rate (MB/sec)
Send/Recv 1 0.000058 0.137356
Send/Recv 2 0.000058 0.275140
Send/Recv 4 0.000059 0.541161
Send/Recv 8 0.000058 1.100435
Send/Recv 16 0.000058 2.201563
Send/Recv 32 0.000075 3.417522
Send/Recv 64 0.000091 5.599606
Send/Recv 128 0.000105 9.746746
Send/Recv 256 0.000146 14.034206
Send/Recv 512 0.000216 18.953094
Send/Recv 1024 0.000427 19.175471
Send/Recv 2048 0.000685 23.914758
Send/Recv 4096 0.001174 27.909333
Send/Recv 8192 0.002143 30.585532
Send/Recv 16384 0.004039 32.451296
Send/Recv 32768 0.007925 33.076855
Send/Recv 65536 0.015577 33.657289
Send/Recv 131072 0.030690 34.167242
Send/Recv 262144 0.061578 34.056645
Send/Recv 524288 0.123510 33.959302
Send/Recv 1048576 0.247028 33.958130
Benchmarking point to point performance with nonblocking operations
Target "pingpong" is up to date.
Kind n time (sec) Rate (MB/sec)
Isend/Irecv 1 0.000075 0.107141
Isend/Irecv 2 0.000074 0.215276
Isend/Irecv 4 0.000075 0.428825
Isend/Irecv 8 0.000075 0.854045
Isend/Irecv 16 0.000076 1.688839
Isend/Irecv 32 0.000086 2.969601
Isend/Irecv 64 0.000100 5.109100
Isend/Irecv 128 0.000118 8.646172
Isend/Irecv 256 0.000161 12.745236
Isend/Irecv 512 0.000236 17.351332
Isend/Irecv 1024 0.000439 18.656344
Isend/Irecv 2048 0.000698 23.476144
Isend/Irecv 4096 0.001180 27.775964
Isend/Irecv 8192 0.002139 30.636467
Isend/Irecv 16384 0.004077 32.146369
Isend/Irecv 32768 0.007985 32.831200
Isend/Irecv 65536 0.015659 33.481015
Isend/Irecv 131072 0.030901 33.933867
Isend/Irecv 262144 0.061962 33.845537
Isend/Irecv 524288 0.123352 34.002838
Isend/Irecv 1048576 0.247212 33.932827
Benchmarking point to point performance with nonblocking operations, head-to-head
Target "pingpong" is up to date.
Kind n time (sec) Rate (MB/sec)
head-to-head Isend/Irecv 1 0.000206 0.077632
head-to-head Isend/Irecv 2 0.000203 0.157965
head-to-head Isend/Irecv 4 0.000198 0.323580
head-to-head Isend/Irecv 8 0.000201 0.636468
head-to-head Isend/Irecv 16 0.000176 1.454645
head-to-head Isend/Irecv 32 0.000174 2.936635
head-to-head Isend/Irecv 64 0.000196 5.228625
head-to-head Isend/Irecv 128 0.000200 10.251717
head-to-head Isend/Irecv 256 0.000254 16.102737
head-to-head Isend/Irecv 512 0.000298 27.515328
head-to-head Isend/Irecv 1024 0.000623 26.286949
head-to-head Isend/Irecv 2048 0.000992 33.034751
head-to-head Isend/Irecv 4096 0.001738 37.711510
head-to-head Isend/Irecv 8192 0.003216 40.759072
head-to-head Isend/Irecv 16384 0.006131 42.756788
head-to-head Isend/Irecv 32768 0.011820 44.356288
head-to-head Isend/Irecv 65536 0.023318 44.967606
head-to-head Isend/Irecv 131072 0.046196 45.397249
head-to-head Isend/Irecv 262144 0.092604 45.292683
head-to-head Isend/Irecv 524288 0.185066 45.327632
head-to-head Isend/Irecv 1048576 0.370367 45.298840
Benchmarking point to point performance with unaligned data
Target "pingpong" is up to date.
Kind char n time (sec) Rate (MB/sec)
Send/Recv 1 0.000059 0.017061
Send/Recv 2 0.000057 0.034862
Send/Recv 4 0.000058 0.069037
Send/Recv 8 0.000057 0.141311
Send/Recv 16 0.000055 0.288501
Send/Recv 32 0.000058 0.554167
Send/Recv 64 0.000056 1.137003
Send/Recv 128 0.000059 2.169755
Send/Recv 256 0.000078 3.293664
Send/Recv 512 0.000091 5.599454
Send/Recv 1024 0.000118 8.678885
Send/Recv 2048 0.000156 13.119793
Send/Recv 4096 0.000218 18.782529
Send/Recv 8192 0.000428 19.134041
Send/Recv 16384 0.000689 23.795798
Send/Recv 32768 0.001190 27.526015
Send/Recv 65536 0.002126 30.829952
Send/Recv 131072 0.004050 32.365854
Send/Recv 262144 0.007905 33.160957
Send/Recv 524288 0.015584 33.643115
Send/Recv 1048576 0.030992 33.833859
Kind double n time (sec) Rate (MB/sec)
Send/Recv 1 0.000059 0.136316
Send/Recv 2 0.000057 0.278667
Send/Recv 4 0.000058 0.549257
Send/Recv 8 0.000058 1.097437
Send/Recv 16 0.000061 2.112267
Send/Recv 32 0.000076 3.357910
Send/Recv 64 0.000093 5.501186
Send/Recv 128 0.000106 9.694346
Send/Recv 256 0.000149 13.790471
Send/Recv 512 0.000223 18.366681
Send/Recv 1024 0.000430 19.056146
Send/Recv 2048 0.000694 23.591498
Send/Recv 4096 0.001190 27.543655
Send/Recv 8192 0.002133 30.726061
Send/Recv 16384 0.004105 31.931495
Send/Recv 32768 0.007982 32.840043
Send/Recv 65536 0.015583 33.645031
Send/Recv 131072 0.030889 33.946995
Send/Recv 262144 0.061654 34.014650
Send/Recv 524288 0.123427 33.982042
Send/Recv 1048576 0.246725 33.999787
Kind int n time (sec) Rate (MB/sec)
Send/Recv 1 0.000059 0.068297
Send/Recv 2 0.000058 0.138879
Send/Recv 4 0.000058 0.276001
Send/Recv 8 0.000059 0.545952
Send/Recv 16 0.000056 1.136393
Send/Recv 32 0.000057 2.233150
Send/Recv 64 0.000073 3.504489
Send/Recv 128 0.000089 5.772847
Send/Recv 256 0.000107 9.570465
Send/Recv 512 0.000160 12.816025
Send/Recv 1024 0.000215 19.036772
Send/Recv 2048 0.000419 19.558315
Send/Recv 4096 0.000683 23.990484
Send/Recv 8192 0.001173 27.925983
Send/Recv 16384 0.002138 30.657249
Send/Recv 32768 0.004090 32.050176
Send/Recv 65536 0.007931 33.054646
Send/Recv 131072 0.015741 33.306472
Send/Recv 262144 0.031088 33.729569
Send/Recv 524288 0.061707 33.985814
Send/Recv 1048576 0.123911 33.849276
Benchmarking point to point performance with contention
Target "pingpong" is up to date.
Kind (np=2) n time (sec) Rate (MB/sec)
Send/Recv 1 0.000058 0.137420
Send/Recv 2 0.000058 0.276431
Send/Recv 4 0.000059 0.541421
Send/Recv 8 0.000059 1.078233
Send/Recv 16 0.000060 2.132502
Send/Recv 32 0.000074 3.460195
Send/Recv 64 0.000089 5.772971
Send/Recv 128 0.000106 9.627452
Send/Recv 256 0.000146 14.063117
Send/Recv 512 0.000220 18.597050
Send/Recv 1024 0.000427 19.180520
Send/Recv 2048 0.000693 23.646401
Send/Recv 4096 0.001180 27.758611
Send/Recv 8192 0.002109 31.073337
Send/Recv 16384 0.004049 32.368651
Send/Recv 32768 0.007851 33.391375
Send/Recv 65536 0.015622 33.561305
Send/Recv 131072 0.031010 33.813647
Send/Recv 262144 0.061716 33.980603
Send/Recv 524288 0.123233 34.035652
Send/Recv 1048576 0.247389 33.908587
Kind (np=4) n time (sec) Rate (MB/sec)
Send/Recv 1 0.000058 0.137785
Send/Recv 2 0.000058 0.273954
Send/Recv 4 0.000058 0.551620
Send/Recv 8 0.000059 1.089568
Send/Recv 16 0.000060 2.119474
Send/Recv 32 0.000078 3.301645
Send/Recv 64 0.000088 5.806635
Send/Recv 128 0.000105 9.769495
Send/Recv 256 0.000146 14.026996
Send/Recv 512 0.000222 18.463962
Send/Recv 1024 0.000429 19.089453
Send/Recv 2048 0.000686 23.892088
Send/Recv 4096 0.001182 27.727195
Send/Recv 8192 0.002136 30.684162
Send/Recv 16384 0.004066 32.236302
Send/Recv 32768 0.007974 32.876080
Send/Recv 65536 0.015565 33.682939
Send/Recv 131072 0.031021 33.802366
Send/Recv 262144 0.061787 33.941769
Send/Recv 524288 0.123614 33.930737
Send/Recv 1048576 0.247025 33.958570
Kind (np=8) n time (sec) Rate (MB/sec)
Send/Recv 1 0.000058 0.138042
Send/Recv 2 0.000059 0.270764
Send/Recv 4 0.000059 0.541327
Send/Recv 8 0.000059 1.088297
Send/Recv 16 0.000061 2.098888
Send/Recv 32 0.000077 3.318992
Send/Recv 64 0.000089 5.737766
Send/Recv 128 0.000108 9.503481
Send/Recv 256 0.000148 13.842512
Send/Recv 512 0.000222 18.449410
Send/Recv 1024 0.000435 18.849520
Send/Recv 2048 0.000691 23.710132
Send/Recv 4096 0.001185 27.645323
Send/Recv 8192 0.002157 30.386990
Send/Recv 16384 0.004074 32.171619
Send/Recv 32768 0.007947 32.985550
Send/Recv 65536 0.015654 33.493046
Send/Recv 131072 0.031118 33.697106
Send/Recv 262144 0.061929 33.863716
Send/Recv 524288 0.123765 33.889193
Send/Recv 1048576 0.246875 33.979204
Kind (np=16) n time (sec) Rate (MB/sec)
Send/Recv 1 0.000063 0.127574
Send/Recv 2 0.000062 0.258132
Send/Recv 4 0.000064 0.497067
Send/Recv 8 0.000062 1.038388
Send/Recv 16 0.000064 2.012623
Send/Recv 32 0.000079 3.252942
Send/Recv 64 0.000091 5.631789
Send/Recv 128 0.000109 9.386038
Send/Recv 256 0.000153 13.377607
Send/Recv 512 0.000223 18.385233
Send/Recv 1024 0.000435 18.817582
Send/Recv 2048 0.000700 23.390260
Send/Recv 4096 0.001190 27.545971
Send/Recv 8192 0.002157 30.376955
Send/Recv 16384 0.004112 31.876552
Send/Recv 32768 0.007993 32.794851
Send/Recv 65536 0.015705 33.383934
Send/Recv 131072 0.031096 33.720390
Send/Recv 262144 0.062040 33.803435
Send/Recv 524288 0.123783 33.884279
Send/Recv 1048576 0.247412 33.905474
Kind (np=32) n time (sec) Rate (MB/sec)
Send/Recv 1 0.000063 0.127404
Send/Recv 2 0.000063 0.255114
Send/Recv 4 0.000065 0.488916
Send/Recv 8 0.000063 1.017032
Send/Recv 16 0.000066 1.944830
Send/Recv 32 0.000080 3.186365
Send/Recv 64 0.000093 5.513281
Send/Recv 128 0.000114 9.007855
Send/Recv 256 0.000160 12.825051
Send/Recv 512 0.000337 12.168292
Send/Recv 1024 0.000497 16.470056
Send/Recv 2048 0.000826 19.828747
Send/Recv 4096 0.001576 20.790559
Send/Recv 8192 0.003011 21.764352
Send/Recv 16384 0.005886 22.267535
Send/Recv 32768 0.011563 22.670394
Send/Recv 65536 0.022851 22.944143
Send/Recv 131072 0.045603 22.993550
Send/Recv 262144 0.090900 23.070903
Send/Recv 524288 0.181638 23.091599
Send/Recv 1048576 0.362657 23.130970
Kind (np=64) n time (sec) Rate (MB/sec)
Send/Recv 1 0.000065 0.122433
Send/Recv 2 0.000066 0.243264
Send/Recv 4 0.000068 0.473078
Send/Recv 8 0.000067 0.951622
Send/Recv 16 0.000066 1.938275
Send/Recv 32 0.000082 3.123656
Send/Recv 64 0.000096 5.357610
Send/Recv 128 0.000113 9.050362
Send/Recv 256 0.000234 8.736580
Send/Recv 512 0.000381 10.743254
Send/Recv 1024 0.000482 16.998936
Send/Recv 2048 0.000900 18.213551
Send/Recv 4096 0.001956 16.754269
Send/Recv 8192 0.003791 17.287658
Send/Recv 16384 0.007645 17.145614
Send/Recv 32768 0.015276 17.160162
Send/Recv 65536 0.030918 16.957385
Send/Recv 131072 0.061523 17.043677
Send/Recv 262144 0.123120 17.033345
Send/Recv 524288 0.245905 17.056586
Send/Recv 1048576 0.492532 17.031591
barrier
Benchmarking collective barrier
Target "barrier" is up to date.
Kind np time (sec)
Barrier 1 0.000009
Barrier 2 0.000236
Barrier 4 0.000430
Barrier 8 0.000657
Barrier 16 0.000900
Barrier 32 0.001151
Barrier 64 0.001446
Benchmarking collective Allreduce
Target "barrier" is up to date.
Kind np time (sec)
Allreduce 1 0.000019
Allreduce 2 0.000260
Allreduce 4 0.000490
Allreduce 8 0.000746
Allreduce 16 0.001020
Allreduce 32 0.001302
Allreduce 64 0.001614
vector
Comparing the performance of MPI vector datatypes
Target "vector" is up to date.
Kind n stride time (sec) Rate (MB/sec)
Vector 1000 24 0.002044 3.914301
Struct 1000 24 0.002679 2.986191
User 1000 24 0.000576 13.891699
User(add) 1000 24 0.000572 13.979156
circulate
Pipelining pitfalls
Target "circulate" is up to date.
For n = 20000, m = 20000, T_comm = 0.006858, T_compute = 0.024717, sum = 0.031575, T_both = 0.030687
For n = 500, m = 500, T_comm = 0.000187, T_compute = 0.000630, sum = 0.000817, T_both = 0.000996
3way
Exploring the cost of synchronization delays
Target "bad" is up to date.
[2] Litsize = 8, Time for first send = 0.000057, for second = 0.000037
[2] Litsize = 9, Time for first send = 0.000067, for second = 0.000038
[2] Litsize = 511, Time for first send = 0.000191, for second = 0.000154
[2] Litsize = 512, Time for first send = 0.000190, for second = 0.000159
[2] Litsize = 513, Time for first send = 0.000349, for second = 0.001796
jacobi
Jacobi Iteration - Example Parallel Mesh
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
send/recv: 25 iterations in 0.051521 secs (0.543472 MFlops); diffnorm 0.036615, m=7 n=34 np=16
send/recv: 25 iterations in 0.903593 secs (25.384875 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
send/recv: 25 iterations in 0.054429 secs (1.028859 MFlops); diffnorm 0.055291, m=7 n=66 np=32
send/recv: 25 iterations in 1.837686 secs (24.963569 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
send/recv: 25 iterations in 0.063062 secs (1.776035 MFlops); diffnorm 0.080560, m=7 n=130 np=64
send/recv: 25 iterations in 3.627254 secs (25.294725 MFlops); diffnorm 0.474303, m=4098 n=130 np=64
Jacobi Iteration - Shift up and down
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
shift/sendrecv: 25 iterations in 0.056335 secs (0.497023 MFlops); diffnorm 0.036615, m=7 n=34 np=16
shift/sendrecv: 25 iterations in 0.182103 secs (125.959467 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
shift/sendrecv: 25 iterations in 0.061043 secs (0.917386 MFlops); diffnorm 0.055291, m=7 n=66 np=32
shift/sendrecv: 25 iterations in 0.203695 secs (225.214708 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
shift/sendrecv: 25 iterations in 1.037380 secs (0.107964 MFlops); diffnorm 0.080560, m=7 n=130 np=64
shift/sendrecv: 25 iterations in 0.197551 secs (464.439409 MFlops); diffnorm 0.474303, m=4098 n=130 np=64
Jacobi Iteration - Exchange head-to-head
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
head-to-head sendrecv: 25 iterations in 0.036205 secs (0.773370 MFlops); diffnorm 0.036615, m=7 n=34 np=16
head-to-head sendrecv: 25 iterations in 0.191498 secs (119.779575 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
head-to-head sendrecv: 25 iterations in 0.045439 secs (1.232428 MFlops); diffnorm 0.055291, m=7 n=66 np=32
head-to-head sendrecv: 25 iterations in 0.217355 secs (211.061386 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
head-to-head sendrecv: 25 iterations in 0.052467 secs (2.134687 MFlops); diffnorm 0.080560, m=7 n=130 np=64
head-to-head sendrecv: 25 iterations in 0.214366 secs (428.008522 MFlops); diffnorm 0.474303, m=4098 n=130 np=64
Jacobi Iteration - Nonblocking send/recv
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
irecv/isend: 25 iterations in 0.037117 secs (0.754363 MFlops); diffnorm 0.036615, m=7 n=34 np=16
irecv/isend: 25 iterations in 0.203043 secs (112.969327 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
irecv/isend: 25 iterations in 0.044261 secs (1.265233 MFlops); diffnorm 0.055291, m=7 n=66 np=32
irecv/isend: 25 iterations in 0.339861 secs (134.982292 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
irecv/isend: 25 iterations in 0.075359 secs (1.486220 MFlops); diffnorm 0.080560, m=7 n=130 np=64
irecv/isend: 25 iterations in 0.232182 secs (395.166074 MFlops); diffnorm 0.474303, m=4098 n=130 np=64
Jacobi Iteration - Nonblocking send/recv for receiver pull
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
isend/irecv: 25 iterations in 0.037104 secs (0.754638 MFlops); diffnorm 0.036615, m=7 n=34 np=16
isend/irecv: 25 iterations in 0.186693 secs (122.862898 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
isend/irecv: 25 iterations in 0.043085 secs (1.299759 MFlops); diffnorm 0.055291, m=7 n=66 np=32
isend/irecv: 25 iterations in 0.249176 secs (184.107286 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
isend/irecv: 25 iterations in 0.051272 secs (2.184420 MFlops); diffnorm 0.080560, m=7 n=130 np=64
isend/irecv: 25 iterations in 0.213290 secs (430.167327 MFlops); diffnorm 0.474303, m=4098 n=130 np=64
Jacobi Iteration - Synchronous send
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
ssend/irecv: 25 iterations in 0.041803 secs (0.669804 MFlops); diffnorm 0.036615, m=7 n=34 np=16
ssend/irecv: 25 iterations in 0.174947 secs (131.111821 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
ssend/irecv: 25 iterations in 0.049397 secs (1.133670 MFlops); diffnorm 0.055291, m=7 n=66 np=32
ssend/irecv: 25 iterations in 0.205755 secs (222.960479 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
ssend/irecv: 25 iterations in 0.064661 secs (1.732104 MFlops); diffnorm 0.080560, m=7 n=130 np=64
ssend/irecv: 25 iterations in 0.195724 secs (468.775594 MFlops); diffnorm 0.474303, m=4098 n=130 np=64
Jacobi Iteration - Ready send
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
rsend: 25 iterations in 0.035526 secs (0.788157 MFlops); diffnorm 0.036615, m=7 n=34 np=16
rsend: 25 iterations in 0.184065 secs (124.616610 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
rsend: 25 iterations in 0.043794 secs (1.278724 MFlops); diffnorm 0.055291, m=7 n=66 np=32
rsend: 25 iterations in 0.206774 secs (221.861388 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
rsend: 25 iterations in 0.351672 secs (0.318478 MFlops); diffnorm 0.080560, m=7 n=130 np=64
rsend: 25 iterations in 0.228262 secs (401.952055 MFlops); diffnorm 0.474303, m=4098 n=130 np=64
Jacobi Iteration - Overlapping communication
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
isend/overlap: 25 iterations in 0.036323 secs (0.770857 MFlops); diffnorm 0.036615, m=7 n=34 np=16
isend/overlap: 25 iterations in 0.211098 secs (108.658512 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
isend/overlap: 25 iterations in 0.045201 secs (1.238902 MFlops); diffnorm 0.055291, m=7 n=66 np=32
isend/overlap: 25 iterations in 0.224026 secs (204.776003 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
isend/overlap: 25 iterations in 0.054266 secs (2.063907 MFlops); diffnorm 0.080560, m=7 n=130 np=64
isend/overlap: 25 iterations in 0.235590 secs (389.449054 MFlops); diffnorm 0.474303, m=4098 n=130 np=64
Jacobi Iteration - Overlapping communication (sends first)
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
send first/overlap: 25 iterations in 0.037683 secs (0.743040 MFlops); diffnorm 0.036615, m=7 n=34 np=16
send first/overlap: 25 iterations in 0.192060 secs (119.429594 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
send first/overlap: 25 iterations in 0.050304 secs (1.113242 MFlops); diffnorm 0.055291, m=7 n=66 np=32
send first/overlap: 25 iterations in 0.219943 secs (208.577558 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
send first/overlap: 25 iterations in 0.051117 secs (2.191057 MFlops); diffnorm 0.080560, m=7 n=130 np=64
send first/overlap: 25 iterations in 0.215668 secs (425.424559 MFlops); diffnorm 0.474303, m=4098 n=130 np=64
Jacobi Iteration - Persistent send/recv
Target "jacobi" is up to date.
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
qsub failed: qsub: Job exceeds queue resource limits
mpirun aborting
persistent send/recv: 25 iterations in 0.035434 secs (0.790209 MFlops); diffnorm 0.036615, m=7 n=34 np=16
persistent send/recv: 25 iterations in 0.199322 secs (115.077927 MFlops); diffnorm 0.468864, m=4098 n=34 np=16
persistent send/recv: 25 iterations in 0.043017 secs (1.301825 MFlops); diffnorm 0.055291, m=7 n=66 np=32
persistent send/recv: 25 iterations in 0.218442 secs (210.011232 MFlops); diffnorm 0.470684, m=4098 n=66 np=32
persistent send/recv: 25 iterations in 0.050107 secs (2.235210 MFlops); diffnorm 0.080560, m=7 n=130 np=64
persistent send/recv: 25 iterations in 0.246237 secs (372.609712 MFlops); diffnorm 0.474303, m=4098 n=130 np=64