(转)《linux性能及调优指南》 3.3 内存瓶颈

翻译：Hank (http://blog.csdn.net/fireroll)
版权所有，尊重他人劳动成果，转载时请注明作者和原始出处及本声明。
原文名称：《Linux Performance and Tuning Guidelines》
原文地址：http://www.redbooks.ibm.com/abstracts/redp4285.html

3.3 内存瓶颈
On a Linux system, many programs run at the same time. 
These programs support multiple users, and some processes are more used than others. 
Some of these programs use a portion of memory while the rest are “sleeping.” 
When an application accesses cache, the performance increases because an in-memory access retrieves data, 
thereby eliminating the need to access slower disks.
在linux系统中，在同一时间有支持多个用户的多个程序在运行，它们对内存使用有多有少；
应该程序访问内存数据的效率要比磁盘高；

The OS uses an algorithm to control which programs will use physical memory and which are paged out. 
This is transparent to user programs. Page space is a file created by the OS on a disk partition 
to store user programs that are not currently in use. Typically, page sizes are 4 KB or 8 KB. 
In Linux, the page size is defined by using the variable EXEC_PAGESIZE in the include/asm-<architecture>/param.h 
kernel header file. The process used to page a process out to disk is called pageout.
操作系统使用一定的策略来决定哪些程序将使用物理内存，哪些程序将被从内存页面中替出；
这些操作对用户程序来说是透明的；
页面空间是由操作系统在磁盘划分上创建的文件，用来存储当前未在使用的用户程序；
通常，页面大小为 4KB 或 8KB; 
在Linux系统中，页面大小由内核的头文件:
include/asm-<architecture>/param.h 
中的宏 EXEC_PAGESIZE 来定义；
进程使用的页面被替换到磁盘称为换页 或 页面溢出；

3.3.1 Finding memory bottlenecks 
寻找内存瓶颈
文章的这部分使用了KDE的一个工具KDE System Guard来做内存检测；

KDE，K桌面环境(Kool Desktop Environment)的缩写。
一种著名的运行于 Linux、Unix 以及FreeBSD 等操作系统上面自由图形工作环境，
整个系统采用的都是 TrollTech 公司所开发的Qt程序库（现在属于Digia公司）。
KDE 和 Gnome 都是 Linux 操作系统上最流行的桌面环境系统。

Start your analysis by listing the applications that are running on the server. 
Determine how much physical memory and swap each application needs to run. 
Figure 3-1 on page 83 shows KDE System Guard monitoring memory usage.

​
图 3.1 KDE System Guard内存监测

​The indicators in Table 3-1 can also help you define a problem with memory.

表3-1

Paging and swapping indicators
换页和交换的意义
In Linux, as with all UNIX-based operating systems, there are differences between paging and swapping. 
Paging moves individual pages to swap space on the disk; 
swapping is a bigger operation that moves the entire address space of a process to swap space in one operation.
在Linux系统中，分布和交换是不同的；
换页是将独立的页面移到磁盘的交换空间；
交换是更大的操作，它将整个进程的地址空间一次性移到磁盘的交换空间；

Swapping can have one of two causes:
导致交换的原因有两个
. A process enters sleep mode. 
  进程进入 sleep 模式
  This usually happens because the process depends on interactive action and editors, shells, 
  and data entry applications spend most of their time waiting for user input. During this time, 
  they are inactive.
  这种情况通常是因为进程需要和编辑器，shell等交互，以及应用程序需要等待用户的数据输入；
  在这是，它是非活动的；

. A process behaves poorly. 
  进程的异常行为
  Paging can be a serious performance problem when the amount of free memory pages falls 
  below the minimum amount specified, because the paging mechanism is not able to handle 
  the requests for physical memory pages and the swap mechanism is called to free more pages. 
  This significantly increases I/O to disk and will quickly degrade a server’s performance.
  当空闲内存页面过小时，换页会导致严重的性能问题，
  因为换页机制不能处理物理内存页面的请求，而交换机制将会调用更多的空闲内存页面。
  这会急剧增加磁盘的I/O, 并快速地拉低服务器的性能

If your server is always paging to disk (a high page-out rate), consider adding more memory.
However, for systems with a low page-out rate, it might not affect performance.
如果服务器总是在换页(一个很高的换页率)，就需要考虑增加内存；
当然，系统换页率低时，它不会影响性能；

3.3.2 Performance tuning options
性能调整选项
It you believe there is a memory bottleneck, 
consider performing one or more of these actions:
如果你确认是内存的瓶颈，可以考虑用下面的办法来解决
. Tune the swap space using bigpages, hugetlb, shared memory.
  调整交换空间，使用大页面，大块内存，共享内存

. Increase or decrease the size of pages.
  增加或降低页面的大小；
​
. Improve the handling of active and inactive memory.
  改进对活动和非活动内存的处理；

. Adjust the page-out rate.
  调整换页率；
​
. Limit the resources used for each user on the server.
  限制服务器上每个用户的可用资源；

. Stop the services that are not needed, as discussed in “Daemons” on page 97.
  停掉不需要的服务

. Add memory.
  增加内存