さくらの[VPS|クラウド]のベンチマーク測定

再起動した後、すぐにUnix Benchmarkで測定した。
さくらのクラウドでは3回、ISOイメージからのインストールのUbuntuの11.04, 10.04.3とテンプレートの10.04.3で測定した。

下が表にしてまとめたもの。

感想

  • Ubuntu 11.04より10.04の方がIndex高くなる希ガス。
  • 単コアはクラウド、2コアはVPS。
  • ディスク速度はクラウド。
  • 本申し込みまで気長に待てるand金がない場合はVPSを。
  • 単コア性能でブイブイ行きたいor切羽詰っているorネットワークを柔軟に構築したいorオートスケールor短期の契約したい場合はクラウドを。

基本、クラウドを使うと良いと思います。

リリース環境として使う場合は、アプリケーションを配備してベンチマークをすると良いと思います。VPSの方だといきなり負荷が上がった場合にすぐに対処できないので、そういった状況へ対応できる点をクラウドとの比較では加味すべきです。

生データ

さくらのVPS

Ubuntu 10.04.3 LTS amd64

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

   System: www28407u: GNU/Linux
   OS: GNU/Linux -- 2.6.32-33-server -- #72-Ubuntu SMP Fri Jul 29 21:21:55 UTC 2011
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Core(TM)2 Duo CPU T7700 @ 2.40GHz (5320.6 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   CPU 1: Intel(R) Core(TM)2 Duo CPU T7700 @ 2.40GHz (5320.6 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   21:51:39 up 0 min,  1 user,  load average: 0.14, 0.04, 0.01; runlevel 2

------------------------------------------------------------------------
Benchmark Run: Fri Nov 18 2011 21:51:39 - 22:19:46
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       23719301.4 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3030.7 MWIPS (9.9 s, 7 samples)
Execl Throughput                               2977.2 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        716634.1 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          202973.8 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1468154.8 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1493949.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 233720.1 lps   (10.0 s, 7 samples)
Process Creation                               3837.4 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   3801.3 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    939.2 lpm   (60.0 s, 2 samples)
System Call Overhead                        3079054.3 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   23719301.4   2032.5
Double-Precision Whetstone                       55.0       3030.7    551.0
Execl Throughput                                 43.0       2977.2    692.4
File Copy 1024 bufsize 2000 maxblocks          3960.0     716634.1   1809.7
File Copy 256 bufsize 500 maxblocks            1655.0     202973.8   1226.4
File Copy 4096 bufsize 8000 maxblocks          5800.0    1468154.8   2531.3
Pipe Throughput                               12440.0    1493949.8   1200.9
Pipe-based Context Switching                   4000.0     233720.1    584.3
Process Creation                                126.0       3837.4    304.6
Shell Scripts (1 concurrent)                     42.4       3801.3    896.5
Shell Scripts (8 concurrent)                      6.0        939.2   1565.4
System Call Overhead                          15000.0    3079054.3   2052.7
                                                                   ========
System Benchmarks Index Score                                        1085.7

------------------------------------------------------------------------
Benchmark Run: Fri Nov 18 2011 22:19:46 - 22:47:53
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       46105320.1 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     6039.4 MWIPS (9.2 s, 7 samples)
Execl Throughput                               5598.9 lps   (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        801118.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          210951.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2183881.2 KBps  (30.0 s, 2 samples)
Pipe Throughput                             3080976.6 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 498331.7 lps   (10.0 s, 7 samples)
Process Creation                              17764.0 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   8093.3 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1059.2 lpm   (60.0 s, 2 samples)
System Call Overhead                        5092618.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   46105320.1   3950.8
Double-Precision Whetstone                       55.0       6039.4   1098.1
Execl Throughput                                 43.0       5598.9   1302.1
File Copy 1024 bufsize 2000 maxblocks          3960.0     801118.3   2023.0
File Copy 256 bufsize 500 maxblocks            1655.0     210951.5   1274.6
File Copy 4096 bufsize 8000 maxblocks          5800.0    2183881.2   3765.3
Pipe Throughput                               12440.0    3080976.6   2476.7
Pipe-based Context Switching                   4000.0     498331.7   1245.8
Process Creation                                126.0      17764.0   1409.8
Shell Scripts (1 concurrent)                     42.4       8093.3   1908.8
Shell Scripts (8 concurrent)                      6.0       1059.2   1765.4
System Call Overhead                          15000.0    5092618.6   3395.1
                                                                   ========
System Benchmarks Index Score                                        1933.5

さくらのクラウド

[パブリック] ubuntu-11.04-server-amd64

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

   System: scloud-test: GNU/Linux
   OS: GNU/Linux -- 2.6.38-12-server -- #51-Ubuntu SMP Wed Sep 28 16:07:08 UTC 2011
   Machine: x86_64 (x86_64)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Xeon(R) CPU X5675 @ 3.07GHz (6133.8 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   21:51:39 up 0 min,  1 user,  load average: 0.00, 0.00, 0.00; runlevel 2

------------------------------------------------------------------------
Benchmark Run: 金 11月 18 2011 21:51:39 - 22:19:45
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       29142907.8 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3553.8 MWIPS (9.9 s, 7 samples)
Execl Throughput                               3572.1 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        988098.7 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          283302.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1534186.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             2022827.0 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 269688.5 lps   (10.0 s, 7 samples)
Process Creation                               9603.6 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4236.4 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    546.8 lpm   (60.1 s, 2 samples)
System Call Overhead                        4032393.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   29142907.8   2497.3
Double-Precision Whetstone                       55.0       3553.8    646.2
Execl Throughput                                 43.0       3572.1    830.7
File Copy 1024 bufsize 2000 maxblocks          3960.0     988098.7   2495.2
File Copy 256 bufsize 500 maxblocks            1655.0     283302.4   1711.8
File Copy 4096 bufsize 8000 maxblocks          5800.0    1534186.0   2645.1
Pipe Throughput                               12440.0    2022827.0   1626.1
Pipe-based Context Switching                   4000.0     269688.5    674.2
Process Creation                                126.0       9603.6    762.2
Shell Scripts (1 concurrent)                     42.4       4236.4    999.1
Shell Scripts (8 concurrent)                      6.0        546.8    911.3
System Call Overhead                          15000.0    4032393.1   2688.3
                                                                   ========
System Benchmarks Index Score                                        1331.2

[パブリック] ubuntu 10.04.3-server-amd64 (テンプレート)

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

   System: test2: GNU/Linux
   OS: GNU/Linux -- 2.6.32-35-server -- #78-Ubuntu SMP Tue Oct 11 16:26:12 UTC 2011
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Xeon(R) CPU X5675 @ 3.07GHz (6132.3 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   22:46:02 up 1 min,  1 user,  load average: 0.00, 0.00, 0.00; runlevel 2

------------------------------------------------------------------------
Benchmark Run: Fri Nov 18 2011 22:46:02 - 23:14:10
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       30216352.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3625.4 MWIPS (9.9 s, 7 samples)
Execl Throughput                               3959.0 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1025784.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          298731.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1980752.2 KBps  (30.0 s, 2 samples)
Pipe Throughput                             2300815.4 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 379467.1 lps   (10.0 s, 7 samples)
Process Creation                              13212.9 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   6071.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    778.5 lpm   (60.1 s, 2 samples)
System Call Overhead                        4229065.7 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   30216352.6   2589.2
Double-Precision Whetstone                       55.0       3625.4    659.2
Execl Throughput                                 43.0       3959.0    920.7
File Copy 1024 bufsize 2000 maxblocks          3960.0    1025784.8   2590.4
File Copy 256 bufsize 500 maxblocks            1655.0     298731.7   1805.0
File Copy 4096 bufsize 8000 maxblocks          5800.0    1980752.2   3415.1
Pipe Throughput                               12440.0    2300815.4   1849.5
Pipe-based Context Switching                   4000.0     379467.1    948.7
Process Creation                                126.0      13212.9   1048.6
Shell Scripts (1 concurrent)                     42.4       6071.2   1431.9
Shell Scripts (8 concurrent)                      6.0        778.5   1297.5
System Call Overhead                          15000.0    4229065.7   2819.4
                                                                   ========
System Benchmarks Index Score                                        1579.9

[パブリック] ubuntu-10.04.3-server-amd64 ISO

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

   System: test3: GNU/Linux
   OS: GNU/Linux -- 2.6.32-35-server -- #78-Ubuntu SMP Tue Oct 11 16:26:12 UTC 2011
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Xeon(R) CPU X5675 @ 3.07GHz (6133.8 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   22:46:07 up 1 min,  1 user,  load average: 0.00, 0.00, 0.00; runlevel 2

------------------------------------------------------------------------
Benchmark Run: 金 11月 18 2011 22:46:07 - 23:14:15
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       29116047.2 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3585.0 MWIPS (9.9 s, 7 samples)
Execl Throughput                               3777.7 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        976042.9 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          303968.9 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1757346.4 KBps  (30.0 s, 2 samples)
Pipe Throughput                             2278245.6 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 357590.6 lps   (10.0 s, 7 samples)
Process Creation                              12605.5 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   5757.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    760.1 lpm   (60.0 s, 2 samples)
System Call Overhead                        4161298.4 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   29116047.2   2494.9
Double-Precision Whetstone                       55.0       3585.0    651.8
Execl Throughput                                 43.0       3777.7    878.5
File Copy 1024 bufsize 2000 maxblocks          3960.0     976042.9   2464.8
File Copy 256 bufsize 500 maxblocks            1655.0     303968.9   1836.7
File Copy 4096 bufsize 8000 maxblocks          5800.0    1757346.4   3029.9
Pipe Throughput                               12440.0    2278245.6   1831.4
Pipe-based Context Switching                   4000.0     357590.6    894.0
Process Creation                                126.0      12605.5   1000.4
Shell Scripts (1 concurrent)                     42.4       5757.2   1357.8
Shell Scripts (8 concurrent)                      6.0        760.1   1266.8
System Call Overhead                          15000.0    4161298.4   2774.2
                                                                   ========
System Benchmarks Index Score                                        1520.9
カテゴリー: VPS, チラシの裏 パーマリンク

コメントを残す

メールアドレスが公開されることはありません。 * が付いている欄は必須項目です

このサイトはスパムを低減するために Akismet を使っています。コメントデータの処理方法の詳細はこちらをご覧ください