vmstat (Virtual Memory Statistics)

vmstat (virtual memory statistics) is a computer system monitoring tool that collects and displays summary information about operating system memory, processes, interrupts, paging and block I/O.

下面是运行 vmstat 的输出例子：

$ vmstat 1 6    # 1表示采集间隔为1秒，6表示共采集6次
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu-----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 1  0      0 4615096 3038056 38104728    0    0     2    32    5    4  6  1 92  0  0
 1  0      0 4595820 3038056 38104728    0    0     0     8 6349 5219 17 25 58  0  0
 3  0      0 4489904 3038056 38106224    0    0     0   308 4182 5663 61 13 25  0  0
 3  0      0 4446832 3038056 38105612    0    0     0   460 3124 3228 78 13  9  0  0
 2  0      0 4388312 3038056 38105612    0    0     0     4 3193 4439 88 12  0  0  0
 2  0      0 4354088 3038056 38105612    0    0     0     0 2817 2873 77  5 18  0  0

默认地（不加参数时）， vmstat 的输出信息分为了 6 组（procs/memory/swap/io/system/cpu），下面分别解释这 6 组报告：

(1) Procs
r: The number of processes waiting for CPU time. (r=running queue)
b: The number of processes in uninterruptible sleep. (b=blocked queue, waiting for resource (e.g. filesystem I/O blocked, inode lock))

Tip 1： 当 r 值大于系统中的 CPU 核数，则往往意味着“CPU 是性能瓶颈”。
Tip 2： b 值表示处于“uninterruptible sleep”状态下的进程数，一般在等待 I/O 时进程会处于“uninterruptible sleep”状态。 当 b 值比较大时，则往往意味着“磁盘是性能瓶颈”。 处于“uninterruptible sleep”状态下的进程，是无法用 kill -9 命令杀掉（因为它不会被信号中断）。处于“uninterruptible sleep”状态的进程在 ps 输出的 STAT 列会被标记为字母 D，如：

$ vmstat                            # 这个例子中有 2 个处于“uninterruptible sleep”状态下的进程
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 0  2      0 189292 691572 10832264    0    0   144   137    0    0  6  0 93  1  0

$ ps aux | awk '$8 ~ /D/'           # 查看所有处于“uninterruptible sleep”状态的进行
root      6708  0.0  0.0      0     0 ?        D    Oct29   0:03 [kworker/u8:3]
root     18506  0.0  0.0      0     0 ?        D    Jun15  24:02 [jbd2/vdc1-8]

(2) Memory
swpd: the amount of virtual memory used.
free: the amount of idle memory.
buff: the amount of memory used as buffers.
cache: the amount of memory used as cache.

Tips：上面 4 项内存相关输出在命令 free 的输出中都可以找到。

(3) Swap
si: Amount of memory swapped in from disk (/s). 每秒由内存进入内存交换区数量。
so: Amount of memory swapped to disk (/s). 每秒由内存交换区进入内存数量。

(4) IO
bi: Blocks received from (read) a block device (blocks/s). 每秒读磁盘的数据。
bo: Blocks sent to (write) a block device (blocks/s). 每秒写磁盘的数据。

(5) System
in: The number of interrupts per second, including the clock.
cs: The number of context switches per second.

(6) CPU
us: Time spent running non-kernel code. (user time, including nice time)
sy: Time spent running kernel code. (system time)
id: Time spent idle. Prior to Linux 2.5.41, this includes IO-wait time.
wa: Time spent waiting for IO. Prior to Linux 2.5.41, included in idle.
st: Time stolen from a virtual machine. Prior to Linux 2.6.11, unknown.

参考：
https://linux.die.net/man/8/vmstat
https://www.thomas-krenn.com/en/wiki/Linux_Performance_Measurements_using_vmstat
http://nonfunctionaltestingtools.blogspot.jp/2013/03/vmstat-output-explained.html