Nodeスペック
Company |
Microsoft |
Product |
Azure Standard F32s_v2 |
CPU |
Platinum 8168 CPU @ 2.70GHz |
Core |
32Core/Node |
Memory |
64GByte |
OS |
Red Hat Enterprise Linux Server release 7.4(Maipo) x64 |
compiler |
gcc バージョン 4.8.5 20150623 (Red Hat 4.8.5-36) (GCC) |
結果性能値
実行時Core数 |
結果値(MFlops) |
1 Core実行時 |
2107.07 |
24 Core実行時 |
25213.79 |
実行結果ログ
single Core 実行結果
################################################
1Core
For example:
Grid-size= XS (32x32x64)
S (64x64x128)
M (128x128x256)
L (256x256x512)
XL (512x512x1024)
Grid-size =
mimax = 128 mjmax = 128 mkmax = 256
imax = 127 jmax = 127 kmax =255
Start rehearsal measurement process.
Measure the performance in 3 times.
MFLOPS: 2107.073046 time(s): 0.195207 1.733593e-03
Now, start the actual measurement process.
The loop will be excuted in 922 times
This will take about one minute.
Wait for a while
uucpu : 58.921465 sec.
Loop executed for 922 times
Gosa : 7.849165e-04
MFLOPS measured : 2145.413991
Score based on Pentium III 600MHz using Fortran 77: 25.898286
################################################
Max-Core OpenMP並列実行結果
################################################
32Core
For example:
Grid-size= XS (32x32x64)
S (64x64x128)
M (128x128x256)
L (256x256x512)
XL (512x512x1024)
Grid-size =
mimax = 128 mjmax = 128 mkmax = 256
imax = 127 jmax = 127 kmax =255
Start rehearsal measurement process.
Measure the performance in 3 times.
MFLOPS: 7793.311759 time(s): 0.052778 1.693367e-03
Now, start the actual measurement process.
The loop will be excuted in 3410 times
This will take about one minute.
Wait for a while
cpu : 18.542569 sec.
Loop executed for 3410 times
Gosa : 2.787795e-04
MFLOPS measured : 25213.793933
Score based on Pentium III 600MHz using Fortran 77: 304.367382
################################################
特記事項
#
#numactl -H
available: 1 nodes (0)
node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
node 0 size: 65535 MB
node 0 free: 352 MB
node distances:
node 0
0: 10
#
フラットに見えてしまうようです。
Post Views: 463