unixbench
1。測出 linux 系統整體的效能
2。結果不只取決於硬體,系統及libraries、compiler都有關系
unixbench 現在在 http://code.google.com/p/byte-unixbench/ 生活著
也已出到 5.1.3 版本了,比 4 版 多了測試顯示的功能
測試的硬體規格:
。Intel(R) Pentium(R) CPU G6950 2.80GHz
。RAM: 2G
。CentOS 5.6
我用 4.1 版及 5.1.3 版測,4 版的分數竟是 0
測試所花的時間快一個小時,真久!
4.1 版的分數
INDEX VALUES
TEST BASELINE RESULT INDEXDhrystone 2 using register variables 116700.0 15441218.3 1323.2
Double-Precision Whetstone 55.0 3055.3 555.5
Execl Throughput 43.0 0.0 0.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 592308.0 1495.7
File Copy 256 bufsize 500 maxblocks 1655.0 177533.0 1072.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 1224604.0 2111.4
Pipe Throughput 12440.0 1180382.9 948.9
Process Creation 126.0 10370.2 823.0
Shell Scripts (8 concurrent) 6.0 1957.0 3261.7
System Call Overhead 15000.0 965449.5 643.6
=========
FINAL SCORE 0.0real 52m10.323s
user 19m51.941s
sys 24m45.498s
5.1.3 版
不知為何有兩個分數,是因為雙核的關系嗎 !?
————————————————————————
Benchmark Run: 六 9月 17 2011 22:12:18 – 22:40:50
2 CPUs in system; running 1 parallel copy of testsDhrystone 2 using register variables 15587038.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3078.7 MWIPS (9.9 s, 7 samples)
Execl Throughput 3924.5 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 594677.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 178827.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1245328.8 KBps (30.0 s, 2 samples)
Pipe Throughput 1197015.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 228417.8 lps (10.0 s, 7 samples)
Process Creation 11341.3 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4149.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1838.3 lpm (60.0 s, 2 samples)
System Call Overhead 979365.3 lps (10.0 s, 7 samples)System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 15587038.5 1335.7
Double-Precision Whetstone 55.0 3078.7 559.8
Execl Throughput 43.0 3924.5 912.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 594677.3 1501.7
File Copy 256 bufsize 500 maxblocks 1655.0 178827.4 1080.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 1245328.8 2147.1
Pipe Throughput 12440.0 1197015.9 962.2
Pipe-based Context Switching 4000.0 228417.8 571.0
Process Creation 126.0 11341.3 900.1
Shell Scripts (1 concurrent) 42.4 4149.9 978.7
Shell Scripts (8 concurrent) 6.0 1838.3 3063.8
System Call Overhead 15000.0 979365.3 652.9
========
System Benchmarks Index Score 1072.0————————————————————————
Benchmark Run: 六 9月 17 2011 22:40:50 – 23:09:25
2 CPUs in system; running 2 parallel copies of testsDhrystone 2 using register variables 32247456.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 6401.8 MWIPS (9.9 s, 7 samples)
Execl Throughput 11189.3 lps (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 331133.8 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 98082.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1078543.2 KBps (30.0 s, 2 samples)
Pipe Throughput 2474684.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching 785465.9 lps (10.0 s, 7 samples)
Process Creation 33406.3 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 14822.3 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 2070.9 lpm (60.0 s, 2 samples)
System Call Overhead 1916200.7 lps (10.0 s, 7 samples)System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 32247456.4 2763.3
Double-Precision Whetstone 55.0 6401.8 1164.0
Execl Throughput 43.0 11189.3 2602.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 331133.8 836.2
File Copy 256 bufsize 500 maxblocks 1655.0 98082.2 592.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 1078543.2 1859.6
Pipe Throughput 12440.0 2474684.8 1989.3
Pipe-based Context Switching 4000.0 785465.9 1963.7
Process Creation 126.0 33406.3 2651.3
Shell Scripts (1 concurrent) 42.4 14822.3 3495.8
Shell Scripts (8 concurrent) 6.0 2070.9 3451.5
System Call Overhead 15000.0 1916200.7 1277.5
========
System Benchmarks Index Score 1809.4real 57m7.387s
user 28m38.594s
sys 36m40.151s
留言