LeeNukes
Linux, technology and various information.
benchmark
HP ProLiant MicroServer Benchmarks
OK, as with my last server I used the nbench CPU benchmark which is pretty long in the tooth now, but as a comparison I ran it again:
BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)
TEST : Iterations/sec. : Old Index : New Index
: : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT : 764.32 : 19.60 : 6.44
STRING SORT : 118 : 52.73 : 8.16
BITFIELD : 1.8813e+08 : 32.27 : 6.74
FP EMULATION : 186.16 : 89.33 : 20.61
FOURIER : 11665 : 13.27 : 7.45
ASSIGNMENT : 14.577 : 55.47 : 14.39
IDEA : 3321.4 : 50.80 : 15.08
HUFFMAN : 1184.8 : 32.86 : 10.49
NEURAL NET : 21.828 : 35.06 : 14.75
LU DECOMPOSITION : 726.08 : 37.61 : 27.16
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX : 43.090
FLOATING-POINT INDEX: 25.960
Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU : Dual AuthenticAMD AMD Athlon(tm) II Neo N36L
Dual-Core Processor 1298MHz
L2 Cache : 1024 KB
OS : Linux 2.6.18-194.32.1.el5
C compiler : gcc version 4.1.2 20080704 (Red Hat 4.1.2-48)
libc : libc-2.5.so
MEMORY INDEX : 9.250
INTEGER INDEX : 12.038
FLOATING-POINT INDEX: 14.398
Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.
I also re-ran the test on my other server:
BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)
TEST : Iterations/sec. : Old Index : New Index
: : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT : 1046.6 : 26.84 : 8.81
STRING SORT : 100.64 : 44.97 : 6.96
BITFIELD : 5.0524e+08 : 86.67 : 18.10
FP EMULATION : 165.2 : 79.27 : 18.29
FOURIER : 18467 : 21.00 : 11.80
ASSIGNMENT : 35.808 : 136.26 : 35.34
IDEA : 3620 : 55.37 : 16.44
HUFFMAN : 1636.1 : 45.37 : 14.49
NEURAL NET : 28.048 : 45.06 : 18.95
LU DECOMPOSITION : 1243.5 : 64.42 : 46.52
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX : 60.118
FLOATING-POINT INDEX: 39.355
Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU : 4 CPU GenuineIntel Intel(R) Xeon(TM) CPU 3.06GHz 3066MHz
L2 Cache : 512 KB
OS : Linux 2.6.18-194.32.1.el5
C compiler : gcc version 4.1.2 20080704 (Red Hat 4.1.2-46)
libc : libc-2.5.so
MEMORY INDEX : 16.452
INTEGER INDEX : 13.998
FLOATING-POINT INDEX: 21.828
Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.
As you can see, its about a third slower but it uses about 3 5ths the power of my last server, and is almost silent.
Sheeva Plug Server Migration?
So, I’m back onto looking at an alternative to my old P3 based server.
I wanted to know the hardware specs of the Sheeva CPU though as even though its 1.2GHz I don’t think its going to be as fast, so I thought I’d check.
I found a page online:
(http://computingplugs.com/index.php/SheevaPlug_Performance)
I had some problems compiling nbench on my desktop, so I had a look around and found one result which helped, it was in Danish, but essentially I had to change this line in the Makefile:
# generic options for gcc
CFLAGS = -s -static -Wall -O3
to this:
# generic options for gcc
CFLAGS = -s -Wall -O3
then ran make and it was fine.
Here are the results from nbench on my server:
BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)
TEST : Iterations/sec. : Old Index : New Index
: : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT : 381.96 : 9.80 : 3.22
STRING SORT : 35.155 : 15.71 : 2.43
BITFIELD : 1.3172e+08 : 22.59 : 4.72
FP EMULATION : 46.504 : 22.31 : 5.15
FOURIER : 7791.5 : 8.86 : 4.98
ASSIGNMENT : 7.2354 : 27.53 : 7.14
IDEA : 1442.2 : 22.06 : 6.55
HUFFMAN : 483.19 : 13.40 : 4.28
NEURAL NET : 9.0077 : 14.47 : 6.09
LU DECOMPOSITION : 364.16 : 18.87 : 13.62
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX : 18.079
FLOATING-POINT INDEX: 13.424
Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU : GenuineIntel Pentium III (Coppermine) 797MHz
L2 Cache : 256 KB
OS : Linux 2.6.18-164.11.1.el5
C compiler : gcc version 4.1.2 20080704 (Red Hat 4.1.2-46)
libc : libc-2.5.so
MEMORY INDEX : 4.343
INTEGER INDEX : 4.642
FLOATING-POINT INDEX: 7.445
Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.
Here are the reults from nbench on my desktop:
BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)
TEST : Iterations/sec. : Old Index : New Index
: : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT : 1270.6 : 32.59 : 10.70
STRING SORT : 276 : 123.32 : 19.09
BITFIELD : 5.0967e+08 : 87.43 : 18.26
FP EMULATION : 276 : 132.44 : 30.56
FOURIER : 26984 : 30.69 : 17.24
ASSIGNMENT : 35.888 : 136.56 : 35.42
IDEA : 7378.1 : 112.85 : 33.50
HUFFMAN : 2718.9 : 75.40 : 24.08
NEURAL NET : 56.11 : 90.14 : 37.91
LU DECOMPOSITION : 1779 : 92.16 : 66.55
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX : 91.588
FLOATING-POINT INDEX: 63.406
Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU : Dual GenuineIntel Intel(R) Core(TM)2 Duo CPU E6750 @ 2.66GHz 2667MHz
L2 Cache : 4096 KB
OS : Linux 2.6.31.9-174.fc12.x86_64
C compiler : gcc version 4.4.2 20091222 (Red Hat 4.4.2-20) (GCC)
libc :
MEMORY INDEX : 23.113
INTEGER INDEX : 22.663
FLOATING-POINT INDEX: 35.167
Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.
Here are the results from nbench on the Sheeva Plug:
BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)
TEST : Iterations/sec. : Old Index : New Index
: : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT : 309.72 : 7.94 : 2.61
STRING SORT : 36.102 : 16.13 : 2.50
BITFIELD : 9.8289e+07 : 16.86 : 3.52
FP EMULATION : 71.075 : 34.10 : 7.87
FOURIER : 356.63 : 0.41 : 0.23
ASSIGNMENT : 4.207 : 16.01 : 4.15
IDEA : 1146.3 : 17.53 : 5.21
HUFFMAN : 452.41 : 12.55 : 4.01
NEURAL NET : 0.49636 : 0.80 : 0.34
LU DECOMPOSITION : 16.116 : 0.83 : 0.60
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX : 15.922
FLOATING-POINT INDEX: 0.646
Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU :
L2 Cache :
OS : Linux 2.6.30-rc1-00002-g1758996
C compiler : gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4)
libc : libc-2.9.so
MEMORY INDEX : 3.317
INTEGER INDEX : 4.549
FLOATING-POINT INDEX: 0.358
Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.
I’m now not so sure if it’s worth upgrading. I’ll have to look into other factors. On the site above which benchmarked the SheevaPlug they state:
“From nbench’s website, the Sheeva Plug falls somewhere around a P3 800Mhz range.”
which is exactly what I’m running in my current Compaq DeskPro Server.
© 2013 LeeNukes. All Rights Reserved.
Powered by Wordpress. Designed by 