お名前.com VPSと物理サーバ(Proliant ML350/Xeon E5506)のベンチを比較してみた

　普通の物理サーバと比較してみる為にベンチを取ってみた。
　ML350(Xeon E5506x1, 12GB RAM, SAS 10000rpm HDD)

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

System: ml350.local: GNU/Linux
OS: GNU/Linux — 2.6.18-194.17.1.el5 — #1 SMP Wed Sep 29 12:50:31 EDT 2010
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap=”UTF-8″, collate=”UTF-8″)
CPU 0: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4264.7 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4264.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 2: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4264.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 3: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4264.6 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
12:09:58 up 3 days, 2:08, 4 users, load average: 1.20, 0.94, 0.68; runlevel

————————————————————————
Benchmark Run: 木 6月 21 2012 12:09:58 – 12:37:56
4 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 12209665.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2426.1 MWIPS (9.8 s, 7 samples)
Execl Throughput 4007.8 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 486551.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 142487.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1155556.7 KBps (30.0 s, 2 samples)
Pipe Throughput 928752.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 270718.7 lps (10.0 s, 7 samples)
Process Creation 11906.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 6160.0 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 2213.6 lpm (60.0 s, 2 samples)
System Call Overhead 751282.6 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 12209665.4 1046.2
Double-Precision Whetstone 55.0 2426.1 441.1
Execl Throughput 43.0 4007.8 932.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 486551.5 1228.7
File Copy 256 bufsize 500 maxblocks 1655.0 142487.1 860.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 1155556.7 1992.3
Pipe Throughput 12440.0 928752.1 746.6
Pipe-based Context Switching 4000.0 270718.7 676.8
Process Creation 126.0 11906.2 944.9
Shell Scripts (1 concurrent) 42.4 6160.0 1452.8
Shell Scripts (8 concurrent) 6.0 2213.6 3689.4
System Call Overhead 15000.0 751282.6 500.9
========
System Benchmarks Index Score 1012.6

————————————————————————
Benchmark Run: 木 6月 21 2012 12:37:56 – 13:06:09
4 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables 47698281.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 9706.0 MWIPS (9.9 s, 7 samples)
Execl Throughput 15605.2 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 350054.4 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 105229.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1028767.1 KBps (30.0 s, 2 samples)
Pipe Throughput 3623665.5 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1243005.6 lps (10.0 s, 7 samples)
Process Creation 47556.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 22027.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 3062.2 lpm (60.0 s, 2 samples)
System Call Overhead 2917520.5 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 47698281.2 4087.3
Double-Precision Whetstone 55.0 9706.0 1764.7
Execl Throughput 43.0 15605.2 3629.1
File Copy 1024 bufsize 2000 maxblocks 3960.0 350054.4 884.0
File Copy 256 bufsize 500 maxblocks 1655.0 105229.8 635.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 1028767.1 1773.7
Pipe Throughput 12440.0 3623665.5 2912.9
Pipe-based Context Switching 4000.0 1243005.6 3107.5
Process Creation 126.0 47556.5 3774.3
Shell Scripts (1 concurrent) 42.4 22027.5 5195.2
Shell Scripts (8 concurrent) 6.0 3062.2 5103.6
System Call Overhead 15000.0 2917520.5 1945.0
========
System Benchmarks Index Score 2446.5

お名前.com VPS 2GBの３パラレルとＭＬ３５０の４パラレルのベンチの比較

　比の数字が大きいほどお名前のインデックスが高い。
　あれ・・・専用サーバよりもＣＰＵ周りの結果が良い。　そして、いつも思っているとおり、ディスク周りが遅くて、物理サーバの半分以下になっている。
　ディスク周りが改善されればパフォーマンス的な不満はほぼ無くなりそうだ。

(354)

Tweet