KAGOYAのTypeCの契約をしたので、どのくらいのスペックなのか、UNIXbenchを利用し、ベンチマークを出してみたいと思います。
まずは、UNIXbenchをインストール。
# yum -y install perl perl-Time-HiRes make gcc # curl https://byte-unixbench.googlecode.com/files/UnixBench5.1.3.tgz | tar zx # cd UnixBench # ./Run
そうすると、60分位？で下記のような結果が返ってきます。
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
------------------------------------------------------------------------
12 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 25878793.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3337.0 MWIPS (9.9 s, 7 samples)
Execl Throughput 2927.9 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 680691.9 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 190079.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1358850.7 KBps (30.0 s, 2 samples)
Pipe Throughput 1594516.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching 119221.5 lps (10.0 s, 7 samples)
Process Creation 6283.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4660.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1420.8 lpm (60.0 s, 2 samples)
System Call Overhead 1844541.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 25878793.3 2217.5
Double-Precision Whetstone 55.0 3337.0 606.7
Execl Throughput 43.0 2927.9 680.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 680691.9 1718.9
File Copy 256 bufsize 500 maxblocks 1655.0 190079.3 1148.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 1358850.7 2342.8
Pipe Throughput 12440.0 1594516.8 1281.8
Pipe-based Context Switching 4000.0 119221.5 298.1
Process Creation 126.0 6283.0 498.7
Shell Scripts (1 concurrent) 42.4 4660.4 1099.2
Shell Scripts (8 concurrent) 6.0 1420.8 2368.0
System Call Overhead 15000.0 1844541.6 1229.7
========
System Benchmarks Index Score 1082.3
------------------------------------------------------------------------
12 CPUs in system; running 12 parallel copies of tests
Dhrystone 2 using register variables 140927733.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 33458.3 MWIPS (9.3 s, 7 samples)
Execl Throughput 20526.3 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 430716.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 108525.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1200886.1 KBps (30.0 s, 2 samples)
Pipe Throughput 8782891.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1269200.9 lps (10.0 s, 7 samples)
Process Creation 51841.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 24212.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 3470.6 lpm (60.1 s, 2 samples)
System Call Overhead 5451121.0 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 140927733.3 12076.1
Double-Precision Whetstone 55.0 33458.3 6083.3
Execl Throughput 43.0 20526.3 4773.6
File Copy 1024 bufsize 2000 maxblocks 3960.0 430716.0 1087.7
File Copy 256 bufsize 500 maxblocks 1655.0 108525.0 655.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 1200886.1 2070.5
Pipe Throughput 12440.0 8782891.4 7060.2
Pipe-based Context Switching 4000.0 1269200.9 3173.0
Process Creation 126.0 51841.1 4114.4
Shell Scripts (1 concurrent) 42.4 24212.6 5710.5
Shell Scripts (8 concurrent) 6.0 3470.6 5784.3
System Call Overhead 15000.0 5451121.0 3634.1
========
System Benchmarks Index Score 3657.4
ちなみに以下はKAGOYAのTypeA（日額30円）のUNIXbenchでのベンチマーク
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
========================================================================
------------------------------------------------------------------------
3 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 29010678.7 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3115.5 MWIPS (9.9 s, 7 samples)
Execl Throughput 3379.0 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 488595.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 126296.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1144297.0 KBps (30.0 s, 2 samples)
Pipe Throughput 1133987.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 97281.5 lps (10.0 s, 7 samples)
Process Creation 8333.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 5011.2 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1374.2 lpm (60.0 s, 2 samples)
System Call Overhead 1627055.3 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 29010678.7 2485.9
Double-Precision Whetstone 55.0 3115.5 566.4
Execl Throughput 43.0 3379.0 785.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 488595.5 1233.8
File Copy 256 bufsize 500 maxblocks 1655.0 126296.3 763.1
File Copy 4096 bufsize 8000 maxblocks 5800.0 1144297.0 1972.9
Pipe Throughput 12440.0 1133987.0 911.6
Pipe-based Context Switching 4000.0 97281.5 243.2
Process Creation 126.0 8333.2 661.4
Shell Scripts (1 concurrent) 42.4 5011.2 1181.9
Shell Scripts (8 concurrent) 6.0 1374.2 2290.3
System Call Overhead 15000.0 1627055.3 1084.7
========
System Benchmarks Index Score 989.9
------------------------------------------------------------------------
Benchmark Run: 月 3月 03 2014 11:57:03 - 12:25:15
3 CPUs in system; running 3 parallel copies of tests
Dhrystone 2 using register variables 84445230.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 9202.1 MWIPS (10.0 s, 7 samples)
Execl Throughput 10424.5 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 457710.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 139064.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1588513.3 KBps (30.0 s, 2 samples)
Pipe Throughput 3420805.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 594929.7 lps (10.0 s, 7 samples)
Process Creation 18852.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 11780.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1657.3 lpm (60.0 s, 2 samples)
System Call Overhead 3651868.3 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 84445230.5 7236.1
Double-Precision Whetstone 55.0 9202.1 1673.1
Execl Throughput 43.0 10424.5 2424.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 457710.3 1155.8
File Copy 256 bufsize 500 maxblocks 1655.0 139064.8 840.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 1588513.3 2738.8
Pipe Throughput 12440.0 3420805.4 2749.8
Pipe-based Context Switching 4000.0 594929.7 1487.3
Process Creation 126.0 18852.4 1496.2
Shell Scripts (1 concurrent) 42.4 11780.9 2778.5
Shell Scripts (8 concurrent) 6.0 1657.3 2762.1
System Call Overhead 15000.0 3651868.3 2434.6
========
System Benchmarks Index Score 2134.1
コメント