自宅にあるミニマシンのベンチマークを確認してみました
こんにちは。
とりあえず、次回以降のネタの下準備のために
まずは自宅にあるミニマシンのスペック・・・というかベンチマークを確認しておきます。
・・・ええ、ぶっちゃけ何故その流れになるかわからないかとは思いますが、
あまりお気になさらず^^;
1.何を使ってベンチマークを取るの?
Linuxサーバのベンチマーク測定ツールとしては有名なunixbenchを使用します。
各マシンへのインストール手順は下記の通り。
// ★↓前提パッケージのインストール↓★ # yum install libXext-devel # yum install freeglut freeglut-devel # yum install perl-Time-HiRes // ★↑前提パッケージのインストール↑★ # wget http://byte-unixbench.googlecode.com/files/UnixBench5.1.3.tgz # tar xvzf UnixBench5.1.3.tgz # cd UnixBench # make
これでインストールが完了します。
どうやら、5.1.2からGooogleCode上に移行しているようです。
2.実行するマシンのスペックは?
2台あるのでその分のスペックを乗せると下記の通りです。
ミニマシン1
- CPU:Core i5 U430(1.20GHz/3072 KB Cache) 2C/4T
- メモリ:PC3-8500(DDR3-1066) 4GB
- ストレージ:500GB HDD
- OS:CentOS6.3 64bit
3.実行結果
UnixBenchをインストールしたディレクトリで下記のコマンドを実行すると
UnixBenchが起動します。下記のメッセージが表示され、計測が開始します。
「sh: 3dinfo: command not found」というエラーは出ましたが、
そもそも画面を持たないマシンのためとりあえず進めます。
# ./Run # # # # # # # ##### ###### # # #### # # # # ## # # # # # # # ## # # # # # # # # # # # ## ##### ##### # # # # ###### # # # # # # ## # # # # # # # # # # # # ## # # # # # # # ## # # # # #### # # # # # ##### ###### # # #### # # Version 5.1.3 Based on the Byte Magazine Unix Benchmark Multi-CPU version Version 5 revisions by Ian Smith, Sunnyvale, CA, USA January 13, 2011 johantheghost at yahoo period com
ミニマシン1結果
------------------------------------------------------------------------ 4 CPUs in system; running 1 parallel copy of tests Dhrystone 2 using register variables 13671342.0 lps (10.0 s, 7 samples) Double-Precision Whetstone 1905.3 MWIPS (9.9 s, 7 samples) Execl Throughput 1097.6 lps (29.7 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 436931.8 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 129861.4 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 994384.2 KBps (30.0 s, 2 samples) Pipe Throughput 809131.9 lps (10.0 s, 7 samples) Pipe-based Context Switching 32962.1 lps (10.0 s, 7 samples) Process Creation 3244.6 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 1875.7 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 745.1 lpm (60.0 s, 2 samples) System Call Overhead 988177.0 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 13671342.0 1171.5 Double-Precision Whetstone 55.0 1905.3 346.4 Execl Throughput 43.0 1097.6 255.2 File Copy 1024 bufsize 2000 maxblocks 3960.0 436931.8 1103.4 File Copy 256 bufsize 500 maxblocks 1655.0 129861.4 784.7 File Copy 4096 bufsize 8000 maxblocks 5800.0 994384.2 1714.5 Pipe Throughput 12440.0 809131.9 650.4 Pipe-based Context Switching 4000.0 32962.1 82.4 Process Creation 126.0 3244.6 257.5 Shell Scripts (1 concurrent) 42.4 1875.7 442.4 Shell Scripts (8 concurrent) 6.0 745.1 1241.8 System Call Overhead 15000.0 988177.0 658.8 ======== System Benchmarks Index Score 550.1 ------------------------------------------------------------------------ 4 CPUs in system; running 4 parallel copies of tests Dhrystone 2 using register variables 26816096.8 lps (10.0 s, 7 samples) Double-Precision Whetstone 4815.0 MWIPS (9.9 s, 7 samples) Execl Throughput 5741.0 lps (29.5 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 335792.0 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 85171.2 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 936318.1 KBps (30.0 s, 2 samples) Pipe Throughput 1462624.6 lps (10.0 s, 7 samples) Pipe-based Context Switching 316827.9 lps (10.0 s, 7 samples) Process Creation 16043.1 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 5960.8 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 797.5 lpm (60.2 s, 2 samples) System Call Overhead 2039949.1 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 26816096.8 2297.9 Double-Precision Whetstone 55.0 4815.0 875.5 Execl Throughput 43.0 5741.0 1335.1 File Copy 1024 bufsize 2000 maxblocks 3960.0 335792.0 848.0 File Copy 256 bufsize 500 maxblocks 1655.0 85171.2 514.6 File Copy 4096 bufsize 8000 maxblocks 5800.0 936318.1 1614.3 Pipe Throughput 12440.0 1462624.6 1175.7 Pipe-based Context Switching 4000.0 316827.9 792.1 Process Creation 126.0 16043.1 1273.3 Shell Scripts (1 concurrent) 42.4 5960.8 1405.9 Shell Scripts (8 concurrent) 6.0 797.5 1329.2 System Call Overhead 15000.0 2039949.1 1360.0 ======== System Benchmarks Index Score 1156.1 ------------------------------------------------------------------------
ミニマシン2結果
------------------------------------------------------------------------ 4 CPUs in system; running 1 parallel copy of tests Dhrystone 2 using register variables 6383301.8 lps (10.0 s, 7 samples) Double-Precision Whetstone 715.7 MWIPS (9.9 s, 7 samples) Execl Throughput 999.5 lps (29.9 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 126970.9 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 34914.0 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 373281.2 KBps (30.0 s, 2 samples) Pipe Throughput 366267.1 lps (10.0 s, 7 samples) Pipe-based Context Switching 28008.7 lps (10.0 s, 7 samples) Process Creation 2490.2 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 1455.4 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 405.3 lpm (60.1 s, 2 samples) System Call Overhead 532920.9 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 6383301.8 547.0 Double-Precision Whetstone 55.0 715.7 130.1 Execl Throughput 43.0 999.5 232.4 File Copy 1024 bufsize 2000 maxblocks 3960.0 126970.9 320.6 File Copy 256 bufsize 500 maxblocks 1655.0 34914.0 211.0 File Copy 4096 bufsize 8000 maxblocks 5800.0 373281.2 643.6 Pipe Throughput 12440.0 366267.1 294.4 Pipe-based Context Switching 4000.0 28008.7 70.0 Process Creation 126.0 2490.2 197.6 Shell Scripts (1 concurrent) 42.4 1455.4 343.3 Shell Scripts (8 concurrent) 6.0 405.3 675.5 System Call Overhead 15000.0 532920.9 355.3 ======== System Benchmarks Index Score 280.9 ------------------------------------------------------------------------ 4 CPUs in system; running 4 parallel copies of tests Dhrystone 2 using register variables 17976368.8 lps (10.0 s, 7 samples) Double-Precision Whetstone 2524.2 MWIPS (9.9 s, 7 samples) Execl Throughput 2755.4 lps (29.9 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 128765.5 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 34052.8 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 410303.0 KBps (30.0 s, 2 samples) Pipe Throughput 605499.1 lps (10.0 s, 7 samples) Pipe-based Context Switching 124486.3 lps (10.0 s, 7 samples) Process Creation 6496.5 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 3160.6 lpm (60.1 s, 2 samples) Shell Scripts (8 concurrent) 426.1 lpm (60.3 s, 2 samples) System Call Overhead 1018378.3 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 17976368.8 1540.4 Double-Precision Whetstone 55.0 2524.2 459.0 Execl Throughput 43.0 2755.4 640.8 File Copy 1024 bufsize 2000 maxblocks 3960.0 128765.5 325.2 File Copy 256 bufsize 500 maxblocks 1655.0 34052.8 205.8 File Copy 4096 bufsize 8000 maxblocks 5800.0 410303.0 707.4 Pipe Throughput 12440.0 605499.1 486.7 Pipe-based Context Switching 4000.0 124486.3 311.2 Process Creation 126.0 6496.5 515.6 Shell Scripts (1 concurrent) 42.4 3160.6 745.4 Shell Scripts (8 concurrent) 6.0 426.1 710.1 System Call Overhead 15000.0 1018378.3 678.9 ======== System Benchmarks Index Score 539.0 ------------------------------------------------------------------------
ベンチマーク結果のうち、前半が1コア辺りのスペック、
後半が全コアを使用した時のスペック・・のようですね。