unixbench是linux下面常用的跑分軟件,用於測試linux性能
下面是Centos7.6測試cpu性能
安裝運行命令
安裝編譯環境
yum install make gcc gcc-c automake autoconf libtool gcc-c++ time perl-Time-HiRes SDL-devel mesa-libGL-devel -y
解壓
tar xzf UnixBench5.1.3.tgz
(說明：unixbench5包含system和graphic測試，如果你需要測試graphic，則需要修改Makefile,不要注釋掉"GRAPHIC_TESTS = defined，同時需要系統提供x11perf命令gl_glibs庫。反之，請將Makefile文件中GRAPHICS_TEST = defined注釋掉，然后編譯運行。)
修改Makefile
cd UnixBench sed -i "s/GRAPHIC_TESTS = defined/#GRAPHIC_TESTS = defined/g" ./Makefile
make安裝
執行測試 ./Run
(如果出現錯誤-bash: ./Run: /usr/bin/perl: bad interpreter: No such file or directory。輸入yum groupinstall "Perl Support"即可)
使用方法：
執行system測試方法：Run
執行graphic測試方法：Run graphics
執行system,graphics測試方法：Run gindex
Run參數用法：
Run [ -q | -v ] [-i <n> ] [-c <n> [-c <n> ...]] [test ...]
選項說明：
-q 不顯示測試過程
-v 顯示測試過程
-i <count> 執行次數，最低3次，默認10
-c <n> 每次測試並行n個copies（並行任務）
對於多cpu系統的性能測試策略，需要統計單任務,多任務及其並行的性能增強。
以4個cpu的PC為例，需要測試兩次，4個CPU就是要並行執行4個copies，
例:Run -q -c 1 -c 4
.............
例如下面是E3-1230v5測試結果
BYTE UNIX Benchmarks (Version 5.1.3)
System: localhost.localdomain: GNU/Linux
OS: GNU/Linux -- 3.10.0-957.el7.x86_64 -- #1 SMP Thu Nov 8 23:39:32 UTC 2018
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E3-1230 v5 @ 3.40GHz (6816.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E3-1230 v5 @ 3.40GHz (6816.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 2: Intel(R) Xeon(R) CPU E3-1230 v5 @ 3.40GHz (6816.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 3: Intel(R) Xeon(R) CPU E3-1230 v5 @ 3.40GHz (6816.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 4: Intel(R) Xeon(R) CPU E3-1230 v5 @ 3.40GHz (6816.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 5: Intel(R) Xeon(R) CPU E3-1230 v5 @ 3.40GHz (6816.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 6: Intel(R) Xeon(R) CPU E3-1230 v5 @ 3.40GHz (6816.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 7: Intel(R) Xeon(R) CPU E3-1230 v5 @ 3.40GHz (6816.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
23:03:05 up 5:38, 3 users, load average: 0.05, 0.22, 2.34; runlevel 5
------------------------------------------------------------------------
Benchmark Run: Sun Apr 25 2021 23:03:05 - 23:32:02
8 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 47330693.0 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3549.9 MWIPS (15.6 s, 7 samples)
Execl Throughput 1332.7 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 450849.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 113806.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1532608.9 KBps (30.0 s, 2 samples)
Pipe Throughput 529575.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 172217.4 lps (10.0 s, 7 samples)
Process Creation 12174.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 6063.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 2934.1 lpm (60.0 s, 2 samples)
System Call Overhead 362424.9 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 47330693.0 4055.8
Double-Precision Whetstone 55.0 3549.9 645.4
Execl Throughput 43.0 1332.7 309.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 450849.1 1138.5
File Copy 256 bufsize 500 maxblocks 1655.0 113806.0 687.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 1532608.9 2642.4
Pipe Throughput 12440.0 529575.4 425.7
Pipe-based Context Switching 4000.0 172217.4 430.5
Process Creation 126.0 12174.7 966.2
Shell Scripts (1 concurrent) 42.4 6063.9 1430.2
Shell Scripts (8 concurrent) 6.0 2934.1 4890.1
System Call Overhead 15000.0 362424.9 241.6
========
System Benchmarks Index Score 944.2 （單核性能）
------------------------------------------------------------------------
Benchmark Run: Sun Apr 25 2021 23:32:02 - 00:00:08
8 CPUs in system; running 8 parallel copies of tests
Dhrystone 2 using register variables 195551422.7 lps (10.0 s, 7 samples)
Double-Precision Whetstone 36210.8 MWIPS (10.3 s, 7 samples)
Execl Throughput 17283.2 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1172264.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 301906.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 3877432.4 KBps (30.0 s, 2 samples)
Pipe Throughput 2583778.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching 836533.6 lps (10.0 s, 7 samples)
Process Creation 62177.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 27820.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 3813.5 lpm (60.1 s, 2 samples)
System Call Overhead 1668708.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 195551422.7 16756.8
Double-Precision Whetstone 55.0 36210.8 6583.8
Execl Throughput 43.0 17283.2 4019.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 1172264.0 2960.3
File Copy 256 bufsize 500 maxblocks 1655.0 301906.8 1824.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 3877432.4 6685.2
Pipe Throughput 12440.0 2583778.3 2077.0
Pipe-based Context Switching 4000.0 836533.6 2091.3
Process Creation 126.0 62177.1 4934.7
Shell Scripts (1 concurrent) 42.4 27820.9 6561.5
Shell Scripts (8 concurrent) 6.0 3813.5 6355.9
System Call Overhead 15000.0 1668708.7 1112.5
========
System Benchmarks Index Score 3982.5 （多核性能）
默認unixbench只支持到16cores，修改參數，取消此限制
sed -i "s/\"System Benchmarks\", 'maxCopies' => 16/\"System Benchmarks\", 'maxCopies' => 0/" Run
sed -i 's/$copies > $maxCopies/$maxCopies > 0 \&\& $copies > $maxCopies/' Run
下面是 AMD EPYC 7282 16-Core Processor (5589.5 bogomips) 跑分
.......
單核跑分：
System Benchmarks Index Score 1332.0
多核跑分:
System Benchmarks Index Score 11523.5
結果僅供參考:
一般單核跑分1000+為優秀, 800+為一般, 500- 為差
參考：
https://blog.csdn.net/weixin_30617561/article/details/95073381
https://blog.csdn.net/qq_36357820/article/details/79959445