linux 软件包：UnixBench 性能测试工具、跑分神器
目录
- unixbench介绍
- unixbench使用
- 测试得分项说明
- Dhrystone 2 using register variables
- Double-Precision Whetstone
- Execl Throughput
- File copy
- Pipe Throughput
- Pipe-based Context Switching
- Process Creation
- System Call Overhead
- Shell Scripts
unixbench介绍
UnixBench是一个类unix系（Unix，BSD，Linux）统下的性能测试工具，一个开源工具，被广泛用与测试linux系统主机的性能。Unixbench的主要测试项目有：系统调用、读写、进程、图形化测试、2D、3D、管道、运算、C库等系统基准性能提供测试数据。
unixbench使用
wget http://soft.laozuo.org/scripts/UnixBench5.1.3.tgz --no-check-certificate
tar xf UnixBench5.1.3.tgz
cd UnixBench
make
./Run # 等待10~60分钟，出结果。
"""
结果如下：
make all
make[1]: Entering directory '/home/dewan/work/UnixBench'
Checking distribution of files
./pgms exists
./src exists
./testdir exists
./tmp exists
./results exists
make[1]: Leaving directory '/home/dewan/work/UnixBench'
sh: 1: 3dinfo: not found
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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
1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
1 x Execl Throughput 1 2 3
1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
1 x File Copy 256 bufsize 500 maxblocks 1 2 3
1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
1 x Process Creation 1 2 3
1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
1 x Shell Scripts (1 concurrent) 1 2 3
1 x Shell Scripts (8 concurrent) 1 2 3
8 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
8 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
8 x Execl Throughput 1 2 3
8 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
8 x File Copy 256 bufsize 500 maxblocks 1 2 3
8 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
8 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
8 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
8 x Process Creation 1 2 3
8 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
8 x Shell Scripts (1 concurrent) 1 2 3
8 x Shell Scripts (8 concurrent) 1 2 3
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: LAPTOP-NFVVPB14: GNU/Linux
OS: GNU/Linux -- 5.10.16.3-microsoft-standard-WSL2 -- #1 SMP Fri Apr 2 22:23:49 UTC 2021
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: AMD Ryzen 5 3500U with Radeon Vega Mobile Gfx (4192.0 bogomips)
Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: AMD Ryzen 5 3500U with Radeon Vega Mobile Gfx (4192.0 bogomips)
Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 2: AMD Ryzen 5 3500U with Radeon Vega Mobile Gfx (4192.0 bogomips)
Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 3: AMD Ryzen 5 3500U with Radeon Vega Mobile Gfx (4192.0 bogomips)
Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 4: AMD Ryzen 5 3500U with Radeon Vega Mobile Gfx (4192.0 bogomips)
Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 5: AMD Ryzen 5 3500U with Radeon Vega Mobile Gfx (4192.0 bogomips)
Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 6: AMD Ryzen 5 3500U with Radeon Vega Mobile Gfx (4192.0 bogomips)
Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 7: AMD Ryzen 5 3500U with Radeon Vega Mobile Gfx (4192.0 bogomips)
Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
21:13:36 up 6:08, 0 users, load average: 0.17, 0.05, 0.01; runlevel unknown
------------------------------------------------------------------------
Benchmark Run: Mon Dec 13 2021 21:13:36 - 21:41:36
8 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 31271617.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone 6893.8 MWIPS (9.9 s, 7 samples)
Execl Throughput 5103.8 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1250688.8 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 353369.6 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2797005.3 KBps (30.0 s, 2 samples)
Pipe Throughput 2190662.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 16276.4 lps (10.0 s, 7 samples)
Process Creation 6255.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 9968.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 3964.2 lpm (60.0 s, 2 samples)
System Call Overhead 2897227.0 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 31271617.9 2679.7
Double-Precision Whetstone 55.0 6893.8 1253.4
Execl Throughput 43.0 5103.8 1186.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 1250688.8 3158.3
File Copy 256 bufsize 500 maxblocks 1655.0 353369.6 2135.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 2797005.3 4822.4
Pipe Throughput 12440.0 2190662.0 1761.0
Pipe-based Context Switching 4000.0 16276.4 40.7
Process Creation 126.0 6255.0 496.4
Shell Scripts (1 concurrent) 42.4 9968.6 2351.1
Shell Scripts (8 concurrent) 6.0 3964.2 6607.0
System Call Overhead 15000.0 2897227.0 1931.5
========
System Benchmarks Index Score 1508.1
------------------------------------------------------------------------
Benchmark Run: Mon Dec 13 2021 21:41:36 - 10:50:25
8 CPUs in system; running 8 parallel copies of tests
Dhrystone 2 using register variables 184612035.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 43384.1 MWIPS (10.1 s, 7 samples)
Execl Throughput 18266.1 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 796475.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 229964.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1808295.6 KBps (30.0 s, 2 samples)
Pipe Throughput 10878971.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1491847.1 lps (10.0 s, 7 samples)
Process Creation 28912.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 36066.7 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 5083.6 lpm (60.0 s, 2 samples)
System Call Overhead 12490140.8 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 184612035.4 15819.4
Double-Precision Whetstone 55.0 43384.1 7888.0
Execl Throughput 43.0 18266.1 4247.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 796475.7 2011.3
File Copy 256 bufsize 500 maxblocks 1655.0 229964.4 1389.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 1808295.6 3117.8
Pipe Throughput 12440.0 10878971.6 8745.2
Pipe-based Context Switching 4000.0 1491847.1 3729.6
Process Creation 126.0 28912.2 2294.6
Shell Scripts (1 concurrent) 42.4 36066.7 8506.3
Shell Scripts (8 concurrent) 6.0 5083.6 8472.6
System Call Overhead 15000.0 12490140.8 8326.8
========
System Benchmarks Index Score 4932.6
"""
测试得分项说明
Dhrystone 2 using register variables
此项用于测试 string handling，因为没有浮点操作，所以深受软件和硬件设计（hardware and software design）、编译和链接（compiler and linker options）、代码优化（code optimazaton）、对内存的cache（cache memory）、等待状态（wait states）、整数数据类型（integer data types）的影响。
Double-Precision Whetstone
这一项测试浮点数操作的速度和效率。这一测试包括几个模块，每个模块都包括一组用于科学计算的操作。覆盖面很广的一系列 c 函数：sin，cos，sqrt，exp，log 被用于整数和浮点数的数学运算、数组访问、条件分支（conditional branch）和程序调用。此测试同时测试了整数和浮点数算术运算。
Execl Throughput
此测试考察每秒钟可以执行的 execl 系统调用的次数。execl 系统调用是 exec 函数族的一员。它和其他一些与之相似的命令一样是 execve（） 函数的前端。
File copy
测试从一个文件向另外一个文件传输数据的速率。每次测试使用不同大小的缓冲区。这一针对文件 read、write、copy 操作的测试统计规定时间（默认是 10s）内的文件 read、write、copy 操作次数。
Pipe Throughput
管道（pipe）是进程间交流的最简单方式，这里的 Pipe throughtput 指的是一秒钟内一个进程可以向一个管道写 512 字节数据然后再读回的次数。需要注意的是，pipe throughtput 在实际编程中没有对应的真实存在。
Pipe-based Context Switching
这个测试两个进程（每秒钟）通过一个管道交换一个不断增长的整数的次数。这一点很向现实编程中的一些应用，这个测试程序首先创建一个子进程，再和这个子进程进行双向的管道传输。
Process Creation
测试每秒钟一个进程可以创建子进程然后收回子进程的次数（子进程一定立即退出）。process creation 的关注点是新进程进程控制块（process control block）的创建和内存分配，即一针见血地关注内存带宽。一般说来，这个测试被用于对操作系统进程创建这一系统调用的不同实现的比较。
System Call Overhead
测试进入和离开操作系统内核的代价，即一次系统调用的代价。它利用一个反复地调用 getpid 函数的小程序达到此目的。
Shell Scripts
测试一秒钟内一个进程可以并发地开始一个 shell 脚本的 n 个拷贝的次数，n 一般取值 1，2，4，8。（我在测试时取 1， 8）。这个脚本对一个数据文件进行一系列的变形操作（transformation）。