新しくVPSを契約したので、お名前.comとさくらで再度ベンチを取ってみた

　仕事関係で新しくお名前.com(KVM)とさくらのVPSサーバーの契約をしたので、それぞれのベンチを更新しておく。

お名前.com(KVM)の1380円コースのUnixBnech

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

System: o2a.studioes.net: GNU/Linux
OS: GNU/Linux — 2.6.18-308.1.1.el5 — #1 SMP Wed Mar 7 04:16:51 EST 2012
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap=”UTF-8″, collate=”UTF-8″)
CPU 0: Intel(R) Core(TM)2 Duo CPU T7700 @ 2.40GHz (6133.4 bogomips)
x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
CPU 1: Intel(R) Core(TM)2 Duo CPU T7700 @ 2.40GHz (6132.4 bogomips)
x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
CPU 2: Intel(R) Core(TM)2 Duo CPU T7700 @ 2.40GHz (6134.8 bogomips)
x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
12:20:41 up 3 days, 19:14, 1 user, load average: 0.09, 0.03, 0.01; runlevel 3

————————————————————————
Benchmark Run: Mon May 28 2012 12:20:41 – 12:48:49
3 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 19161530.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3797.3 MWIPS (9.8 s, 7 samples)
Execl Throughput 1738.7 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 988507.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 289959.6 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1526394.9 KBps (30.0 s, 2 samples)
Pipe Throughput 2619601.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching 311905.9 lps (10.0 s, 7 samples)
Process Creation 6949.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4281.3 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1423.9 lpm (60.0 s, 2 samples)
System Call Overhead 4632913.8 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 19161530.5 1641.9
Double-Precision Whetstone 55.0 3797.3 690.4
Execl Throughput 43.0 1738.7 404.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 988507.3 2496.2
File Copy 256 bufsize 500 maxblocks 1655.0 289959.6 1752.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 1526394.9 2631.7
Pipe Throughput 12440.0 2619601.7 2105.8
Pipe-based Context Switching 4000.0 311905.9 779.8
Process Creation 126.0 6949.5 551.5
Shell Scripts (1 concurrent) 42.4 4281.3 1009.7
Shell Scripts (8 concurrent) 6.0 1423.9 2373.2
System Call Overhead 15000.0 4632913.8 3088.6
========
System Benchmarks Index Score 1346.0

————————————————————————
Benchmark Run: Mon May 28 2012 12:48:49 – 13:16:57
3 CPUs in system; running 3 parallel copies of tests

Dhrystone 2 using register variables 56858986.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 11326.2 MWIPS (10.0 s, 7 samples)
Execl Throughput 13070.2 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 163497.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 50706.7 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 407272.4 KBps (30.0 s, 2 samples)
Pipe Throughput 7740792.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1557688.3 lps (10.0 s, 7 samples)
Process Creation 39957.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 16021.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 2878.0 lpm (60.0 s, 2 samples)
System Call Overhead 7840433.7 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 56858986.2 4872.2
Double-Precision Whetstone 55.0 11326.2 2059.3
Execl Throughput 43.0 13070.2 3039.6
File Copy 1024 bufsize 2000 maxblocks 3960.0 163497.5 412.9
File Copy 256 bufsize 500 maxblocks 1655.0 50706.7 306.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 407272.4 702.2
Pipe Throughput 12440.0 7740792.9 6222.5
Pipe-based Context Switching 4000.0 1557688.3 3894.2
Process Creation 126.0 39957.6 3171.2
Shell Scripts (1 concurrent) 42.4 16021.8 3778.7
Shell Scripts (8 concurrent) 6.0 2878.0 4796.7
System Call Overhead 15000.0 7840433.7 5227.0
========
System Benchmarks Index Score 2279.9

　相変わらずウェットストーンが高い。
　トランスコーダを走らせても、20fps前後出るので自宅に変換用PCを買うのがあほらしくなるスピードだ（Walkmanプロファイルだと100fps超えるし）

さくら2GB(v3) 1480円のUnixBench

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

System: www3363uf.sakura.ne.jp: GNU/Linux
OS: GNU/Linux — 2.6.32-220.17.1.el6.x86_64 — #1 SMP Wed May 16 00:01:37 BST 2012
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap=”UTF-8″, collate=”UTF-8″)
CPU 0: Intel(R) Xeon(R) CPU E5645 (4800.2 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
CPU 1: Intel(R) Xeon(R) CPU E5645 (4800.2 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
CPU 2: Intel(R) Xeon(R) CPU E5645 (4800.2 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
13:08:05 up 6 days, 50 min, 1 user, load average: 0.41, 0.20, 0.12; runlevel 3

————————————————————————
Benchmark Run: Mon May 28 2012 13:08:05 – 13:36:03
3 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 20218009.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2702.1 MWIPS (7.5 s, 7 samples)
Execl Throughput 1405.1 lps (29.3 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 571262.0 KBps (30.1 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 164788.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1143543.4 KBps (30.0 s, 2 samples)
Pipe Throughput 1416847.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 218994.1 lps (10.0 s, 7 samples)
Process Creation 3338.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2728.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 782.9 lpm (60.0 s, 2 samples)
System Call Overhead 3000141.3 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 20218009.8 1732.5
Double-Precision Whetstone 55.0 2702.1 491.3
Execl Throughput 43.0 1405.1 326.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 571262.0 1442.6
File Copy 256 bufsize 500 maxblocks 1655.0 164788.5 995.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 1143543.4 1971.6
Pipe Throughput 12440.0 1416847.9 1138.9
Pipe-based Context Switching 4000.0 218994.1 547.5
Process Creation 126.0 3338.4 264.9
Shell Scripts (1 concurrent) 42.4 2728.8 643.6
Shell Scripts (8 concurrent) 6.0 782.9 1304.8
System Call Overhead 15000.0 3000141.3 2000.1
========
System Benchmarks Index Score 881.0

————————————————————————
Benchmark Run: Mon May 28 2012 13:36:03 – 14:04:01
3 CPUs in system; running 3 parallel copies of tests

Dhrystone 2 using register variables 60631625.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 8157.5 MWIPS (7.7 s, 7 samples)
Execl Throughput 7662.9 lps (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 406180.4 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 120266.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1120928.0 KBps (30.0 s, 2 samples)
Pipe Throughput 4245141.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching 581530.1 lps (10.0 s, 7 samples)
Process Creation 16977.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 8931.4 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 1361.9 lpm (60.1 s, 2 samples)
System Call Overhead 6741630.9 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 60631625.5 5195.5
Double-Precision Whetstone 55.0 8157.5 1483.2
Execl Throughput 43.0 7662.9 1782.1
File Copy 1024 bufsize 2000 maxblocks 3960.0 406180.4 1025.7
File Copy 256 bufsize 500 maxblocks 1655.0 120266.1 726.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 1120928.0 1932.6
Pipe Throughput 12440.0 4245141.3 3412.5
Pipe-based Context Switching 4000.0 581530.1 1453.8
Process Creation 126.0 16977.0 1347.4
Shell Scripts (1 concurrent) 42.4 8931.4 2106.5
Shell Scripts (8 concurrent) 6.0 1361.9 2269.9
System Call Overhead 15000.0 6741630.9 4494.4
========
System Benchmarks Index Score 1940.0

　さくらとお名前.com(KVM)のUnixBenchの比。
　File周りでさくらVPSは2倍以上優速だけど、その他はドライストーンが微妙に勝っている以外、ほとんど負けている。
　DB処理でディスクアクセスが多発する場合はさくら、オンメモリで動作しちゃうような場合はお名前の方が良い感じかな。

(266)

Tweet