ubuntu 8.10 amd64 64bit bench mark

amd 64 でテストしてみた結果

hdbench clone

無理やりコンパイルしてみた。

hdd でテスト。上と比べるとやはりライトが大分違いますが、読み込みはキャッシュが効いてる!

Super PI 104万桁 http://pi2.cc.u-tokyo.ac.jp/index-j.htmlftp://pi.super-computing.org/Linux_jp/

コンパイル済みのものを使っているからそんなに変わらない?

$ ./super_pi 20
 Version 2.0 of the super_pi for Linux OS
 Fortran source program was translated into C program with version 19981204 of
 f2c, then generated C source program was optimized manually.
 pgcc 3.2-3 with compile option of "-fast -tp px -Mbuiltin -Minline=size:1000 -Mnoframe -Mnobounds -Mcache_align -Mdalign -Mnoreentrant" was used for the
 compilation.
 ------ Started super_pi run : 2008年 11月 3日 月曜日 14:41:29 JST
 Start of PI calculation up to 1048576 decimal digits
 End of initialization. Time=       0.384 Sec.
 I= 1 L=       0        Time=       1.208 Sec.
 I= 2 L=       0        Time=       1.388 Sec.
 I= 3 L=       1        Time=       1.384 Sec.
 I= 4 L=       2        Time=       1.388 Sec.
 I= 5 L=       5        Time=       1.384 Sec.
 I= 6 L=      10        Time=       1.384 Sec.
 I= 7 L=      21        Time=       1.388 Sec.
 I= 8 L=      43        Time=       1.384 Sec.
 I= 9 L=      87        Time=       1.388 Sec.
 I=10 L=     174        Time=       1.384 Sec.
 I=11 L=     349        Time=       1.384 Sec.
 I=12 L=     698        Time=       1.384 Sec.
 I=13 L=    1396        Time=       1.384 Sec.
 I=14 L=    2794        Time=       1.380 Sec.
 I=15 L=    5588        Time=       1.380 Sec.
 I=16 L=   11176        Time=       1.372 Sec.
 I=17 L=   22353        Time=       1.360 Sec.
 I=18 L=   44707        Time=       1.316 Sec.
 I=19 L=   89415        Time=       1.232 Sec.
 End of main loop
 End of calculation.    Time=      27.350 Sec.
 End of data output.    Time=       0.136 Sec.
 Total calculation(I/O) time=      27.486(       0.744) Sec.
 ------ Ended super_pi run : 2008年 11月 3日 月曜日 14:41:56 JST

姫野ベンチ http://w3cic.riken.go.jp/HPC/HimenoBMT/index.html

少し速くなっている。

$ make
gcc -c -O3 -DSMALL himenoBMTxps.c
gcc -o bmt himenoBMTxps.o -O3 -DSMALL

./bmt
mimax = 65 mjmax = 65 mkmax = 129
imax = 64 jmax = 64 kmax =128
 Start rehearsal measurement process.
 Measure the performance in 3 times.

 MFLOPS: 1219.558320 time(s): 0.040509 3.288628e-03

 Now, start the actual measurement process.
 The loop will be excuted in 4443 times
 This will take about one minute.
 Wait for a while

 Loop executed for 4443 times
 Gosa : 2.522595e-06 
 MFLOPS measured : 1309.829037	cpu : 55.859178
 Score based on Pentium III 600MHz : 15.973525

cpuinfo

微妙に周波数が変わっている。

model name	: AMD Athlon(tm) 64 X2 Dual Core Processor 5000+
stepping	: 2
cpu MHz		: 2611.998
cache size	: 512 KB