MediaTemple GS UnixBench跑分

前几天用简单的命令测试了GS的性能,那时只测试了下载速度、磁盘性能等,今天抽空给MediaTemple GS做了个UnixBench跑分测试。从跑分测试结果来看,性能还是蛮强劲的。

如果你使用的是全新的主机,那么你需要安装几个模块:

yum install make
yum install gcc
yum install perl
yum install perl-Time-HiRes

安装好模块以后,然后再执行下面的操作:

#wget https://byte-unixbench.googlecode.com/files/UnixBench5.1.3.tgz
#tar -zxvf UnixBench5.1.3.tgz
#cd UnixBench
#make
#./Run

敲完这段命令,等半个小时......如果你也在使用MediaTemple GS,也可以测试一下。因为是集群,分配到的服务器繁忙程度不一样,跑分上会有差别,我测试了两次,分配到两台不同的服务器上,跑分差10%左右。总体来说524.8的分数还算不错了。

========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)

System: n19: GNU/Linux
OS: GNU/Linux -- 3.2.6mtv13 -- #1 SMP Tue May 14 15:32:13 PDT 2013
Machine: x86_64 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU: no details available
; runlevel

------------------------------------------------------------------------
Benchmark Run: Wed Jan 08 2014 23:49:57 - 00:20:48
unknown CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 13100159.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2653.1 MWIPS (10.1 s, 7 samples)
Execl Throughput 1708.8 lps (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 454557.9 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 157844.9 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 471958.9 KBps (30.0 s, 2 samples)
Pipe Throughput 478073.2 lps (10.0 s, 7 samples)
Pipe-based Context Switching 168639.1 lps (10.0 s, 7 samples)
Process Creation 3294.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1073.5 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 247.7 lpm (60.2 s, 2 samples)
System Call Overhead 772062.0 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 13100159.8 1122.6
Double-Precision Whetstone 55.0 2653.1 482.4
Execl Throughput 43.0 1708.8 397.4
File Copy 1024 bufsize 2000 maxblocks 3960.0 454557.9 1147.9
File Copy 256 bufsize 500 maxblocks 1655.0 157844.9 953.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 471958.9 813.7
Pipe Throughput 12440.0 478073.2 384.3
Pipe-based Context Switching 4000.0 168639.1 421.6
Process Creation 126.0 3294.4 261.5
Shell Scripts (1 concurrent) 42.4 1073.5 253.2
Shell Scripts (8 concurrent) 6.0 247.7 412.8
System Call Overhead 15000.0 772062.0 514.7
========
System Benchmarks Index Score 524.8

了解更多关于UnixBench