ServerSpeed Messen - Mit Ergebnissen

[carsten]

Hallo zusammen,

ich wollte einen Thread starten um die Geschwindigkeit von dedizierten Servern messbar zu machen. Standardmäßig in vielen Foren verwendet wird dafür UnixBench.

Das ist recht einfach installierbar:

wget http://byte-unixbench.googlecode.com...h-5.1.2.tar.gz
gunzip unixbench-5.1.2.tar.gz
tar xvf unixbench-5.1.2.tar
cd unixbench-5.1.2

Dann empfehle ich die graphischen Tests (setzen X11 voraus) zu deaktivieren. Dies geht in dem ihr im Makefile die Anweisung GRAPHIC_TESTS = defined mittels eines vorangestellten # auskommentiert.

Dann noch:

make

aufrufen und der Test ist fertig gebaut. (Ggf. müssen hierfür die Build Tools installiert werden. (Unter Debian: apt-get install make gcc g++ time file)

Der Test ausgeführt wird dann mit

./Run dhry2reg whetstone-double syscall pipe context1 spawn execl shell1 shell8 shell16

Testet das gerne einmal aus und postet hier dann die Ergebnisse Eurer Hardware.

Grüße,

Carsten

[carsten]

Und hier die Ergebnisse für einen Hetzner Server:

BYTE UNIX Benchmarks (Version 5.1.2)

System: www.webhostlist.de: GNU/Linux
OS: GNU/Linux -- 2.6.32-5-amd64 -- #1 SMP Mon Jan 16 16:22:28 UTC 2012
Machine: x86_64 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 2: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.2 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 3: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 4: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 5: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 6: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 7: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 8: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 9: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 10: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 11: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz (6683.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
11:11:48 up 2 days, 1:07, 1 user, load average: 0.10, 0.68, 0.41; runlevel 2

------------------------------------------------------------------------
Benchmark Run: Tue Mar 27 2012 11:11:48 - 11:35:55
12 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 33947052.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4368.6 MWIPS (9.9 s, 7 samples)
Execl Throughput 2426.0 lps (29.9 s, 2 samples)
Pipe Throughput 2909663.2 lps (10.0 s, 7 samples)
Pipe-based Context Switching 83719.7 lps (10.0 s, 7 samples)
Process Creation 10379.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 5459.5 lpm (60.0 s, 2 samples)
Shell Scripts (16 concurrent) 3608.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 6110.9 lpm (60.0 s, 2 samples)
System Call Overhead 5026910.4 lps (10.0 s, 7 samples)

System Benchmarks Partial Index BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 33947052.3 2908.9
Double-Precision Whetstone 55.0 4368.6 794.3
Execl Throughput 43.0 2426.0 564.2
Pipe Throughput 12440.0 2909663.2 2339.0
Pipe-based Context Switching 4000.0 83719.7 209.3
Process Creation 126.0 10379.1 823.7
Shell Scripts (1 concurrent) 42.4 5459.5 1287.6
Shell Scripts (16 concurrent) --- 3608.5 ---
Shell Scripts (8 concurrent) 6.0 6110.9 10184.9
System Call Overhead 15000.0 5026910.4 3351.3
========
System Benchmarks Index Score (Partial Only) 1417.5

------------------------------------------------------------------------
Benchmark Run: Tue Mar 27 2012 11:35:55 - 12:00:06
12 CPUs in system; running 12 parallel copies of tests

Dhrystone 2 using register variables 212866939.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 41425.9 MWIPS (10.0 s, 7 samples)
Execl Throughput 44585.4 lps (29.9 s, 2 samples)
Pipe Throughput 19816737.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 4116282.5 lps (10.0 s, 7 samples)
Process Creation 121720.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 69894.3 lpm (60.0 s, 2 samples)
Shell Scripts (16 concurrent) 4534.2 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 9089.8 lpm (60.0 s, 2 samples)
System Call Overhead 17919794.4 lps (10.0 s, 7 samples)

System Benchmarks Partial Index BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 212866939.3 18240.5
Double-Precision Whetstone 55.0 41425.9 7532.0
Execl Throughput 43.0 44585.4 10368.7
Pipe Throughput 12440.0 19816737.4 15929.9
Pipe-based Context Switching 4000.0 4116282.5 10290.7
Process Creation 126.0 121720.7 9660.4
Shell Scripts (1 concurrent) 42.4 69894.3 16484.5
Shell Scripts (16 concurrent) --- 4534.2 ---
Shell Scripts (8 concurrent) 6.0 9089.8 15149.6
System Call Overhead 15000.0 17919794.4 11946.5
========
System Benchmarks Index Score (Partial Only) 12359.6