【Linux知识】服务器性能测试--UnixBench，性能测试--unixbench
链接地址： http://blog.csdn.net/jason_asia/article/details/38309079
1.1. 服务器性能测试UnixBench
分别在DELL R720和DELL R610作为测试的服务器，因服务器配置不同，结果仅供参考.
下载：
wget http://byte-unixbench.googlecode.com/files/UnixBench5.1.3.tgz
安装：
tar zxf UnixBench5.1.3.tgz
cd UnixBench
make
运行：
./Run
服务器配置信息：
Product Name: Dell Inc. PowerEdge R720
Host Name: test
Kernel Version: 2.6.32-358.el6.x86_64 x86_64
OS Version: CentOS release 6.4 (Final)
Time Zone: Asia/Ho_Chi_Minh +0700
Char Set: en_US
Memory: 4 memories,total size is 32768MB,Speed: 1600 MHz,Type: DDR3
CPU info: Intel(R) Xeon(R) CPU E5-2609 v2 @2.50GHz, 2 * 4 core, cache size 10240 KB
Swap Info: 7999 MB, vm.swappiness = 0
Disk Capacity: /dev/sda: 584.7 GB
Raid Info: PERC H710 Mini, 512MB cache,BBU,FW21.2.0-0007
RAID1, strip 64 KB, 4 * 2disks/span
Current Cache Policy:WriteBack, ReadAdaptive, Direct, Write Cache OK if Bad BBU
BBU Info: Next Learn time: Tue May 27 17:40:472014,Relative State of Charge: 83 %,Charger Status: Unknown,isSOHGood: Yes
Disk Info: 8 * 136.732 GB , 6.0Gb/s ,port speed6.0Gb/s
File System: /home: xfs ,527G
/tmp: tmpfs ,16G
I/O scheduler: sda [deadline]
测试结果：
# time ./Run
make all
make[1]: Entering directory`/root/UnixBench'
Checking distribution of files
./pgms exists
./src exists
./testdir exists
./tmp exists
./results exists
make[1]: Leaving directory`/root/UnixBench'
sh: 3dinfo: command 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 registervariables 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 2000maxblocks 1 2 3
1 x File Copy 256 bufsize 500maxblocks 1 2 3
1 x File Copy 4096 bufsize 8000maxblocks 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 registervariables 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 2000maxblocks 1 2 3
8 x File Copy 256 bufsize 500maxblocks 1 2 3
8 x File Copy 4096 bufsize 8000maxblocks 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: vn_dpcq_backup_dbserver: GNU/Linux
OS: GNU/Linux -- 2.6.32-358.el6.x86_64 -- #1 SMP Fri Feb 22 00:31:26 UTC2013
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8",collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E5-2609 v2 @ 2.50GHz (4999.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E5-2609 v2 @ 2.50GHz (4999.3 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 2: Intel(R) Xeon(R) CPU E5-2609 v2 @ 2.50GHz (4999.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU3: Intel(R) Xeon(R) CPU E5-2609 v2 @ 2.50GHz (4999.3 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 4: Intel(R) Xeon(R) CPU E5-2609 v2 @ 2.50GHz (4999.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 5: Intel(R) Xeon(R) CPU E5-2609 v2 @ 2.50GHz (4999.3 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 6: Intel(R) Xeon(R) CPU E5-2609 v2 @ 2.50GHz (4999.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 7: Intel(R) Xeon(R) CPU E5-2609 v2 @ 2.50GHz (4999.3 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
22:00:22 up 77 days, 10:34, 2users, load average: 1.15, 1.01, 0.93;runlevel 3
------------------------------------------------------------------------
Benchmark Run: Wed Jul 30 2014 22:00:22 -22:27:47
8 CPUs in system; running 1 parallel copyof tests
Dhrystone 2 using register variables 27570040.0 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3041.3 MWIPS (6.5 s, 7samples)
Execl Throughput 1855.2 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 816467.6 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 218024.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1536957.9 KBps (30.0 s, 2 samples)
Pipe Throughput 1454276.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 88740.9 lps (10.0 s, 7 samples)
Process Creation 4974.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 3078.0 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1663.2 lpm (60.0 s, 2 samples)
System Call Overhead 1912485.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 27570040.0 2362.5
Double-Precision Whetstone 55.0 3041.3 553.0
Execl Throughput 43.0 1855.2 431.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 816467.6 2061.8
File Copy 256 bufsize 500 maxblocks 1655.0 218024.2 1317.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 1536957.9 2649.9
Pipe Throughput 12440.0 1454276.0 1169.0
Pipe-based Context Switching 4000.0 88740.9 221.9
Process Creation 126.0 4974.0 394.8
Shell Scripts (1 concurrent) 42.4 3078.0 725.9
Shell Scripts (8 concurrent) 6.0 1663.2 2772.1
System Call Overhead 15000.0 1912485.4 1275.0
========
System Benchmarks Index Score 1005.3
------------------------------------------------------------------------
Benchmark Run: Wed Jul 30 2014 22:27:47 -22:55:01
8 CPUs in system; running 8 parallel copiesof tests
Dhrystone 2 using register variables 221136112.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 24325.0 MWIPS (6.0 s, 7samples)
Execl Throughput 30167.6 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 281771.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 77252.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 878007.3 KBps (30.0 s, 2 samples)
Pipe Throughput 11413805.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 2370561.4 lps (10.0 s, 7 samples)
Process Creation 74692.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 35687.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 5028.6 lpm (60.0 s, 2 samples)
System Call Overhead 4188710.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 221136112.5 18949.1
Double-Precision Whetstone 55.0 24325.0 4422.7
Execl Throughput 43.0 30167.6 7015.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 281771.5 711.5
File Copy 256 bufsize 500 maxblocks 1655.0 77252.4 466.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 878007.3 1513.8
Pipe Throughput 12440.0 11413805.1 9175.1
Pipe-based Context Switching 4000.0 2370561.4 5926.4
Process Creation 126.0 74692.8 5928.0
Shell Scripts (1 concurrent) 42.4 35687.9 8417.0
Shell Scripts (8 concurrent) 6.0 5028.6 8381.0
System Call Overhead 15000.0 4188710.7 2792.5
========
System Benchmarks Index Score 4037.6
real 54m39.103s
user 57m37.687s
sys 95m44.095s
#################################
服务器配置信息：
Product Name: Dell Inc. PowerEdge R610
Host Name: test
Kernel Version: 2.6.18-194.el5 x86_64
OS Version: Red Hat Enterprise Linux Serverrelease 5.3 (Tikanga)
Time Zone: Asia/Shanghai +0800
Char Set: en_US
Memory: 8 memories,total size is 16384MB,Speed: 1333 MHz (0.8 ns),Type: <OUT OF SPEC>
CPU info: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz, 2 * 4 core, cache size 4096 KB
Swap Info: 8189 MB, vm.swappiness = 0
Disk Capacity: /dev/sda: 146.1 GB
Raid Info: PERC H700 Integrated, 512MBcache,BBU,FW 12.10.1-0001
RAID1, strip 64 KB, 1* 2 disks/span
Current Cache Policy:WriteBack, ReadAdaptive, Direct, Write Cache OK if Bad BBU
BBU Info: Next Learn time: Fri Aug1 06:17:162014,Relative State of Charge: 99 %,,
Disk Info: 2 * 136.732 GB , 6.0Gb/s ,port speed6.0Gb/s
File System: /home: xfs ,109G
/tmp: tmpfs ,7.9G
Nic Info: eth0 2.0.8e,link detectedis yes,speed 1000Mb/s.
eth1 2.0.8e,link detectedis yes,speed 1000Mb/s.
I/O scheduler: sda [deadline]
测试结果：
# ./Run
make all
make[1]: Entering directory`/root/UnixBench'
Checking distribution of files
./pgms exists
./src exists
./testdir exists
./tmp exists
./results exists
make[1]: Leaving directory`/root/UnixBench'
sh: 3dinfo: command 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 registervariables 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 2000maxblocks 1 2 3
1 x File Copy 256 bufsize 500maxblocks 1 2 3
1 x File Copy 4096 bufsize 8000maxblocks 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 registervariables 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 2000maxblocks 1 2 3
8 x File Copy 256 bufsize 500maxblocks 1 2 3
8 x File Copy 4096 bufsize 8000maxblocks 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: my_tlbb_9018_192.168.104.124: GNU/Linux
OS: GNU/Linux -- 2.6.18-194.el5 -- #1 SMP Tue Mar 16 21:52:39 EDT 2010
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8",collate="UTF-8")
CPU0: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4255.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 2: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4256.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 3: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 4: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4256.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 5: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 6: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4256.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
CPU 7: Intel(R) Xeon(R) CPU E5506 @ 2.13GHz (4256.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT,SYSCALL/SYSRET, Intel virtualization
21:44:42 up 3:48, 2 users, load average: 0.19, 0.43, 0.24; runlevel 3
------------------------------------------------------------------------
Benchmark Run: Wed Jul 30 2014 21:44:42 -22:12:14
8 CPUs in system; running 1 parallel copyof tests
Dhrystone 2 using register variables 12114116.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2427.0 MWIPS (8.0 s, 7samples)
Execl Throughput 2141.1 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 688694.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 218516.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1317519.2 KBps (30.0 s, 2 samples)
Pipe Throughput 1782756.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching 170764.9 lps (10.0 s, 7 samples)
Process Creation 8493.3 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4612.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 2277.8 lpm (60.0 s, 2 samples)
System Call Overhead 2958948.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 12114116.1 1038.1
Double-Precision Whetstone 55.0 2427.0 441.3
Execl Throughput 43.0 2141.1 497.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 688694.5 1739.1
File Copy 256 bufsize 500 maxblocks 1655.0 218516.2 1320.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 1317519.2 2271.6
Pipe Throughput 12440.0 1782756.3 1433.1
Pipe-based Context Switching 4000.0 170764.9 426.9
Process Creation 126.0 8493.3 674.1
Shell Scripts (1 concurrent) 42.4 4612.5 1087.9
Shell Scripts (8 concurrent) 6.0 2277.8 3796.3
System Call Overhead 15000.0 2958948.5 1972.6
========
System Benchmarks Index Score 1122.4
------------------------------------------------------------------------
Benchmark Run: Wed Jul 30 2014 22:12:14 -22:39:50
8 CPUs in system; running 8 parallel copiesof tests
Dhrystone 2 using register variables 96804820.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 17210.6 MWIPS (8.3 s, 7samples)
Execl Throughput 24850.3 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 372899.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 107051.9 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 899185.2 KBps (30.0 s, 2 samples)
Pipe Throughput 14107729.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 3443772.2 lps (10.0 s, 7 samples)
Process Creation 55426.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 26969.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 5222.1 lpm (60.0 s, 2 samples)
System Call Overhead 4752771.2 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 96804820.8 8295.2
Double-Precision Whetstone 55.0 17210.6 3129.2
Execl Throughput 43.0 24850.3 5779.1
File Copy 1024 bufsize 2000 maxblocks 3960.0 372899.1 941.7
File Copy 256 bufsize 500 maxblocks 1655.0 107051.9 646.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 899185.2 1550.3
Pipe Throughput 12440.0 14107729.1 11340.6
Pipe-based Context Switching 4000.0 3443772.2 8609.4
Process Creation 126.0 55426.6 4398.9
Shell Scripts (1 concurrent) 42.4 26969.8 6360.8
Shell Scripts (8 concurrent) 6.0 5222.1 8703.6
System Call Overhead 15000.0 4752771.2 3168.5
========
System Benchmarks Index Score 3852.0
linux 系统管理，linux 网络服务，linux 安全，数据库等等，关于编程最好会一点，这主要根据企业要求。关于网络最好也要会一点。反正做运维接触面一点要广。
Linux是一个主要运行于inetel架构的UNIX-like操作系统，它不仅具备UNIX系统的全部特征，而且与POSIX标准兼容。Linux的功能包括真正的多任务、虚拟内存、共享库、需求装载、共享的写时复制程序执行、优秀的内存管理以及TCP/IP网络支持等。它的发行遵守GNU的通用公共许可证。而Windows 操作系统却和Linux有着完全不同的开发模式和发展历史。
进程管理及调度：Linux开销较小
在进程管理及调度方面，Linux是一个单块式的操作系统，操作系统通常在用户进程的内存空间内进行，可免去发生系统调用时的进程切换开销。Windows是一个准微内核操作系统，许多功能以单独的进程实现，从而提高了系统的模块化程度，但进程切换上的开销要大一些。
Linux支持内核级的线程，但它将线程定义为进程的另一个“执行上下文”，从而简化了进程/线程之间的关系和调度程序的设计，它的线程库提供了和POSIX兼容的线程同步机制。Windows支持的线程是典型的进程/线程模型。它的线程同步机制和Linux类似。Linux中进程是基本的调度单位，它提供了三种调度策略。
进程间通信机制：Linux更灵活
在进程间通信机制上，Linux提供了标准的UNIX IPC机制，而Windows 则在基本IPC机制的基础上，提供了许多直接面向应用程序的高级IPC机制。从本质上看，Linux的IPC机制接近于IPC原语，比较底层，提供了最大的灵活性，也可以在此基础上建立更加复杂的高级IPC机制，而Windows 在原语级提供的IPC机制有管道、命名管道、消息传递、共享内存、信号量等。
内存管理：Linux优于Windows
在内存管理上，Linux 和Windows 在面对相同的进程地址空间大小时，对内存布局的使用方式不同。Windows实际只为进程准备了2GB弱的可用虚拟地址空间，而Linux中的进程地址空间使用更灵活些。
在虚拟内存的使用上，Linux内核为用户管理了非常多的细节问题，用户可以认为自己真正拥有4GB地址空间，而不用关心虚拟内存是否提交物理存储等问题。Windows为程序员提供了多种虚拟内存使用方案，虽然这些方案的使用有些复杂，但却提供了一定程度上的灵活性。它也支持DOS和Win16程序的执行，但为了提供这种兼容性，它的内存管理付出了极高的性能代价。
Linux 和Windows虽然均提供了内存共享技术，但它们的实现有些细微的差别。Linux提供给用户的接口非常简单，只需将自己的虚拟内存空间区域附加到共享内存对象之上。Windows则是通过内存映射文件提供共享内存机制，从使用上讲，略显复杂。
Linux的内存交换管理灵活性很强，用户可以在普通的文件系统上建立“无洞”的文件作为交换空间，还可以使用多个交换文件，从而可以动态增加交换文件。它也提供了利用交换分区作为交换空间的方法，此方法是优选的交换空间方案。Windows 的页面文件很难摆脱碎片化的危险，为了保证它采用无碎片的页面文件，必须采取一定的措施。
Linux 和Windows 在高速缓存的使用上有一些共同之处，均尽量采用系统的全部空闲RAM作为高速缓存区域，但Linux在高速缓存管理上有一些独到之处，这是它性能之所以高的根本原因所在。
安全性：Linux具有一定优势
在安全性方面，Linux采用的是UNIX在安全性方面成功的技术，尽管有一些安全漏洞，但因为设计上的开放性，这些漏洞能够在很快的时间内发现并得到解决。Windows ......余下全文>>