System Management: View CPU, memory, disk, I/O, load, performance status

View CPU Related information :cat/proc/cpuinfo

• Processor: Logical CPU serial number starting from 0

• Physical ID: Physical CPU serial number

• CPU Cores: The current physical CPU's core number, according to physical ID can be judged

• Cat/proc/cpuinfo | grep "Physical ID" | Sort | Uniq | Wc-l

#物理cpu个数

• Cat/proc/cpuinfo | grep "Processor" | Wc-l

#逻辑cpu个数 (equals all the cores of all physical CPUs added)

View Memory Usage: Free

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M02/6F/72/wKiom1WcugzAuehtAABhpxvCp-I541.jpg "style=" float: none; "title=" 2015-06-22_10-53-55.png "alt=" Wkiom1wcugzauehtaabhpxvcp-i541.jpg "/>

• Buffers:buffer Cache size of disk caches for disk block reads and writes

• cached:cached cache size to read and write to file Inode

• -buffers/cache: The amount of memory actually used by the program (used-buffers-cached)

• +buffers/cache: Number of available memory (Free+buffers+cache)

#即被挪用的内存数看 +buffers/cache is OK.

Generally, as long as swap swap space is not used, it means that the server memory is sufficient

To view your hard disk usage :

View disk and partition information: Fdisk-l

To view file system Information: DF

• DF [-m displays size in megabytes] [-K displays size in kilobytes] [-H in gb/mb/kb way] [-I with inode quantity display] [-A lists all file systems including special systems such as/proc] [-t display file system type] [-H displays in 1024=1000 mode]

• DF reads the data inside the Superblock

View a directory size: Du

• du  [-s Totals only] [-s does not include statistics for subdirectories] [-A lists all statistics, including files] [-C last plus totals] [-h] [-K] [-m]

• application: du-s * | Sort-rn | Head-n #列出当前目录下最大的10个目录或文件

[Email protected] ~]# Du-s * | Sort-rn | Head

38928 httpd-2.4.12

13448 apr-1.5.2

12580 apr-util-1.5.4

4940 httpd-2.4.12.tar.bz2

1008 apr-1.5.2.tar.gz

856 apr-util-1.5.4.tar.gz

Install.log

8 Install.log.syslog

4 INDEX.HTML.1

4 index.html

To view the I/O performance of a hard disk: Iostat-x-K 1 3 #每秒一次共三次

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M00/6F/6F/wKioL1Wcu9vQYR2cAAE6nyx-iLI452.jpg "style=" float: none; "title=" 2015-06-25_14-18-30.png "alt=" Wkiol1wcu9vqyr2caae6nyx-ili452.jpg "/>

Note the following two points

• %util: The percentage of I/O operations in a second, that is, how much time in one second I/O queues are non-empty, and if close to%100 indicates too many I/O requests, overloads, disk bottlenecks

• await: average per Device I/O operation wait time, generally should be less than 5ms (reference from the book), if greater than 5ms indicates disk I/O pressure, should upgrade hardware or tuning

View System Average Load

When the server response is slow, it may be a load problem, you should look at the average system load, that is, the average CPU load, to see if there are a large number of processes in the queue, the characteristic time of the average number of processes running queues can reflect the system's busy level

Uptime

• Load average: The average number of processes in the process queue in the last 1 minutes, 5 minutes, 15 minutes, these three values are generally not larger than the number of system logical CPUs, and occasionally out of the question

Top: The content of the first line is the content of uptime

• You can also view process priorities, CPU usage, memory usage, and more

View the overall system performance status

Vmstat: A comprehensive system performance analysis tool that allows you to observe the status of the process, memory usage, virtual memory usage, disk I/O, interrupts, context switches, CPU usage, and more. It can also be used as a monitoring standard tool when performing server performance tests.

650) this.width=650; "src=" Http://s3.51cto.com/wyfs02/M02/6F/6F/wKioL1Wcuu-C34g0AADmXHtgmkk048.jpg "title=" 2015-06-22_10-53-55.png "alt=" Wkiol1wcuu-c34g0aadmxhtgmkk048.jpg "/>

• Procs

R: Number of processes waiting to run, how many processes are divided into CPUs, generally not more than the number of CPUs is the normal value

B: Number of processes in non-disruptive sleep state, that is, the number of processes waiting on resource allocation, blocking state

• Memory

SWPD: Virtual memory usage (KB)

Free: Idle memory (KB)

Buff: The amount of memory used as the cache, the cache is the basic contents of the file directory, the location on disk, permissions, etc. (KB)

Cache: The number of memory used as the file cache, caching open files, improving execution efficiency and performance (KB)

• Swap

Si: Number of swap pages from disk to memory, that is, the amount of virtual memory used per second (KB/S)

So: number of swap pages from memory swap to disk (KB/S)

• Io

BI: The number of blocks (blocks/s) sent to a block device that is generally disk

Bo: Number of blocks received from the Block device (block/s)

• System

In: Interrupts per second, including clock interrupts

CS: Number of context switches per second

• CPU (Displays the total CPU usage percentage)

US: User CPU Usage time

SY: System CPU usage time, such as I/O operation, etc.

ID: Idle Time

WA: CPU time waiting for I/O, typically 0

• General us+sy+id=100, because my test virtual machine is not running a task, so both US and Sy are 0

• General US+SY<70 indicates that the system performance is better, if more than 85 may need to check

This article is from the "Call Me boxin" blog, so be sure to keep this source http://boxinknown.blog.51cto.com/10435935/1671996

System Management: View CPU, memory, disk, I/O, load, performance status