http://blog.csdn.net/for_tech/article/details/53044610

CentOS 6及以前

$ free              total       used        free    shared    buffers    cachedMem:        4040360    4012200       28160         0     176628   3571348-/+ buffers/cache:      264224     3776136Swap:       4200956      12184     4188772$

内存的使用分作4部分：

A. 程序使用的；

B. 未被分配的；

C. Buffers (buffer cache)

D. Cached (page cache)

显然，A (程序使用的) 肯定是used，B (未被分配的) 肯定是free。但是，C (Buffers) 和 D (Cached) 是算作used还是算作free呢？一方面，它们已经被分配了，可以算作used；另一方面，当程序需要时，可以回收它们来使用，可以算作free。所以，怎么算都合理。

这就是在free命令的output中，第一行和第二行的区别：

第一行(Mem)：Buffers和Cached被算作used。也就是说，它的free是指 B (未被分配的)；它的used是指 A ＋ C ＋ D；

第二行(-/+ buffers/cache)：Buffers和Cached被算作free。也就是说，它的used是指 A (程序使用的)；它的free是指 B + C + D；行名称“-/+ buffers/cache”的含义就是“把Buffers和Cached从used减下来，加到free里”。

搞清这些之后，我们可以算出A，B，C和D各自的值：

A＝264224

B＝28160

C＝176628

D＝3571348

可见验证一下：

total＝A ＋ B ＋ C ＋ D

第一行used ＝ A ＋ C ＋ D

第二行free ＝ B ＋ C ＋ D

CentOS 7

free命令的out：

total        used        free      shared  buff/cache   availableMem:        1012952      252740      158732       11108      601480      543584Swap:       1048572        5380     1043192

首先，C (Buffers) 和D (Cached)被和到一起，即buff/cache;

其次，used就是指A (程序使用的)；free就是指B (未被分配的)；

另外，CentOS 7中加入了一个available，它是什么呢？手册上是这么说的：

MemAvailable: An estimate of how much memory is available for starting new applications, without swapping.

前面说过，

当程序需要时，可以回收C (Buffers)和D (Cached)，那么MemAvailabe不就是B+C+D吗？当程序需要时可以回收C和D，这句话以前是正确，但是现在就不精确了：因为，现在，C和D中不是所有的内存都可以被回收。所以，

大致可以这么理解，MemAvailable ＝ B (未被分配的) + C (Buffers) ＋ D (Cached) - 不可回收的部分。哪些不可回收呢？共享内存段，tmpfs，ramfs等。详细的介绍如下：

/proc/meminfo: provide estimated available memory

Many load balancing and workload placing programs check /proc/meminfo to estimate how much free memory

is available. They generally do this by adding up "free" and "cached", which was fine ten years ago,

but is pretty much guaranteed to be wrong today.

It is wrong because Cached includes memory that is not freeable as page cache, for example shared memory

segments, tmpfs, and ramfs, and it do esnot include reclaimable slab memory, which can take up a large

fraction of system memory on mostly idle systems with lots of files.

Currently, the amount of memory that is available for a new workload,without pushing the system into swap,

can be estimated from MemFree, Active(file), Inactive(file), and SReclaimable, as well as the "low"watermarks

from /proc/zoneinfo.

However, this may change in the future, and user space really should not be expected to know kernel internals

to come up with an estimate for the amount of free memory.

It is more convenient to provide such an estimate in /proc/meminfo. If things change in the future, we only

have to change it in one place.

你可能感兴趣的文章