Resultados do teste LINPACK (gina-n2 com HT)

Informações sobre o Benchmark

Thu Oct 21 19:49:16 BRST 2010
Intel(R) LINPACK data

Current date/time: Thu Oct 21 19:49:16 2010

CPU frequency:    2.667 GHz
Number of CPUs: 24
Number of threads: 24

Parameters are set to:

Number of tests                             : 15
Number of equations to solve (problem size) : 1000  2000  5000  10000 15000 18000 20000 22000 25000 26000 27000 30000 35000 40000 45000
Leading dimension of array                  : 1000  2000  5008  10000 15000 18008 20016 22008 25000 26000 27000 30000 35000 40000 45000
Number of trials to run                     : 4     2     2     2     2     2     2     2     2     2     1     1     1     1     1    
Data alignment value (in Kbytes)            : 4     4     4     4     4     4     4     4     4     4     4     1     1     1     1    

Maximum memory requested that can be used = 16200901024, at the size = 45000

Performance Summary (GFlops) / per computer (gina-n2)

Size LDA Align. Average Maximal
1000 1000 4 29.0048 32.1607
2000 2000 4 41.7022 41.7060
5000 5008 4 48.6168 48.6214
10000 10000 4 47.0850 47.1401
15000 15000 4 49.4608 49.4889
18000 18008 4 49.7802 49.7851
20000 20016 4 52.4325 52.4475
22000 22008 4 52.5112 52.5321
25000 25000 4 52.4307 52.4307
26000 26000 4 52.8933 52.9074
27000 27000 4 52.9904 52.9904
30000 30000 1 52.9861 52.9861
35000 35000 1 53.0009 53.0009
40000 40000 1 43.6002 43.6002
45000 45000 1 44.0688 44.0688

Thu Oct 21 21:28:51 BRST 2010

Resultados do teste LINPACK (gina SEM HT)

Informações sobre o Benchmark

Thu Oct 21 21:47:08 BRST 2010
Intel(R) LINPACK data

Current date/time: Thu Oct 21 21:47:08 2010

CPU frequency:    2.666 GHz
Number of CPUs: 12
Number of threads: 12

Parameters are set to:

Number of tests                             : 15
Number of equations to solve (problem size) : 1000  2000  5000  10000 15000 18000 20000 22000 25000 26000 27000 30000 35000 40000 45000
Leading dimension of array                  : 1000  2000  5008  10000 15000 18008 20016 22008 25000 26000 27000 30000 35000 40000 45000
Number of trials to run                     : 4     2     2     2     2     2     2     2     2     2     1     1     1     1     1    
Data alignment value (in Kbytes)            : 4     4     4     4     4     4     4     4     4     4     4     1     1     1     1    

Maximum memory requested that can be used = 16200901024, at the size = 45000

Performance Summary (GFlops)

Size LDA Align. Average Maximal
1000 1000 4 58.4767 68.1379
2000 2000 4 88.7300 88.8663
5000 5008 4 107.6973 107.8838
10000 10000 4 115.2268 115.2456
15000 15000 4 123.9998 124.0858
18000 18008 4 125.2082 125.2181
20000 20016 4 124.3444 124.7159
22000 22008 4 126.2919 126.3162
25000 25000 4 126.2928 126.9063
26000 26000 4 127.5811 127.5965
27000 27000 4 127.9242 127.9242
30000 30000 1 128.0746 128.0746
35000 35000 1 128.3174 128.3174
40000 40000 1 129.7266 129.7266
45000 45000 1 130.0088 130.0088

Thu Oct 21 22:30:22 BRST 2010

Resultados do teste LINPACK (MPI)

Resultados do benchmark Himeno

C

(single thread, compilation gcc -O3)

c_simple | Large problem | Per core: 2.50 Gflops

c_simple | Medium problem | Per core: 2.65 Gflops

c_simple | Small problem | Per core: 2.63 Gflops

(single thread, compilation gcc -O3 -march=native -msse4.2)

c_simple | Large problem | Per core: 2.77 Gflops

c_simple | Medium problem | Per core: 2.72 Gflops

c_simple | Small problem | Per core: 2.69 Gflops

(single thread, compilation icc -axSSE4.2)

c_simple | Large problem | Per core: 3.90 Gflops

(single thread, compilation icc -fast)

c_simple | Large problem | Per core: 3.97 Gflops

(automatic multithreaded, compilation icc -fast -parallel)

c_simple | Large problem | Total: 11.40 Gflops

FORTRAN

(single thread, compilation ifort )

c_simple | Large problem | Per core: 4.40 Gflops

c_simple | Medium problem | Per core: 5.04 Gflops

c_simple | Small problem | Per core: 5.06 Gflops

(single thread, compilation ifort -fast)

c_simple | Large problem | Per core: 5.21 Gflops

c_simple | Medium problem | Per core: 5.15 Gflops

c_simple | Small problem | Per core: 4.99 Gflops

(single thread, compilation ifort -fast -msse4.1 -funroll-loops -unroll-aggressive)

c_simple | Large problem | Per core: 5.36 Gflops

c_simple | Medium problem | Per core: 5.28 Gflops

c_simple | Small problem | Per core: 5.20 Gflops

(single thread, compilation ifort -fast -msse4.1 -funroll-loops -unroll-aggressive -fp-model strict)

c_simple | Large problem | Per core: 2.75 Gflops

c_simple | Medium problem | Per core: 2.71 Gflops

c_simple | Small problem | Per core: 2.72 Gflops

 
cpu_intel_xeon_x5650.txt · Última modificação: 2010/10/22 09:44 por algol
 
Exceto onde for informado ao contrário, o conteúdo neste wiki está sob a seguinte licença:CC Attribution-Noncommercial-Share Alike 3.0 Unported
Recent changes RSS feed Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki