vmstat是Virtual Meomory Statistics（虚拟内存统计）的缩写，可 对操作系统的虚拟内存、进程、CPU活动进行监视。它是对系统的整体 情况进行统计，不足之处是无法对某个进程进行深入分析。可以用来 分析系统性能瓶颈。
vmstat reports information about processes, memory, paging, block IO, traps, disks and cpu activity. These reports are intended to help identify system bottlenecks. Linux vmstat does not count itself as a running process.
Optionsvmstat [-V]列出版本信息: -Vjustin@Justin:~$ vmstat -V procps version 3.2.8
vmstat [-a] [-n] [delay [ count]]一般用法，vmstat的语法如下：
vmstat [-V] [-n] [delay [count]]
其中，－n表示在周期性循环输出时，输出的头部信息仅显示一 次；delay是两次输出之间的延迟时间；count是指按照这个时间间隔 统计的次数（默认无限）。对于vmstat输出各字段的含义，可运行 man vmstat查看
delay is the delay between updates in seconds. If no delay is specified, only one report is printed with the average values since boot.
count is the number of updates. If no count is specified and delay is defined, count defaults to infinity.
命令多数情况下输出的是一段时间内的统计值，但是也有一些累加值：
-njustin@Justin:~$ vmstat -n 3 procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa 0 0 1376 65960 39056 376444 0 0 248 24 333 1601 12 11 72 4 0 0 1376 65216 39056 376444 0 0 0 1 388 1206 2 3 95 0 0 0 1376 72352 39056 376444 0 0 0 0 377 1108 7 5 89 0 ^C
-aThe -a switch displays active/inactive memory, given a
justin@Justin:~$ vmstat -a procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free inact active si so bi bo in cs us sy id wa 0 0 1376 71448 452636 451496 0 0 247 24 334 1605 12 11 72
no optionsjustin@Justin:~$ vmstat procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa 0 0 0 22024 34488 510808 0 0 1305 45 342 1460 22 18 35 25
vmstat [-f] [-s] [-m]-fThe -f switch displays the number of forks since boot. This includes the fork, vfork, and clone system calls, and is equivalent to the total number of tasks created. Each process is represented by one or more tasks, depending on thread usage. This display does not repeat.
tasks = fork, vfork (followed by exec), clone (thread)
The display is accumulated since boot.
justin@Justin:~$ vmstat -f
3658 forks
justin@Justin:~$ vmstat -f
3659 forks
justin@Justin:~$ vmstat -f
3660 forks
-sThis display does not repeat.The -s switch displays a table of various event counters and memory statistics.
vmstat -v
justin@Justin:~$ vmstat -s
1010128 K total memory ; total physical memory
952140 K used memory ; used
457848 K active memory
458760 K inactive memory ; to be swap out
57988 K free memory ; free used+free=total
78568 K buffer memory
406676 K swap cache
979956 K total swap ; total swap space
524 K used swap ; used swap space
979432 K free swap ; free swap space used+free = total
25333 non-nice user cpu ticks
238 nice user cpu ticks
25509 system cpu ticks
132072 idle cpu ticks
15898 IO-wait cpu ticks
404 IRQ cpu ticks
313 softirq cpu ticks
0 stolen cpu ticks
913509 pages paged in
74632 pages paged out
32 pages swapped in
142 pages swapped out
655728 interrupts
3233646 CPU context switches
1330667648 boot time
3865 forks
Active memory is currently being used by a process. Inactive memory is memory that is allocated but has not been used for a while. Nothing is essentially different between the two types of memory. When required, the Linux kernel takes a process’s least recently used memory pages and moves them from the active to the inactive list. When choosing which memory will be swapped to disk, the kernel chooses from the inactive memory list.
-mFIELD DESCRIPTION FOR SLAB MODE
The display is accumulated since boot.cache: Cache name num: Number of currently active objects total: Total number of available objects size: Size of each object pages: Number of pages with at least one active object
The -m displays slabinfo.
cat /proc/slabinfo
cat /proc/slabinfoslabinfo - version: 2.1 # name <active_objs> <num_objs> <objsize> <objperslab> <pagesperslab> : tunables <limit> <batchcount> <sharedfactor> : slabdata <active_slabs> <num_slabs> <sharedavail> VMBlockInodeCache 7 7 4480 7 8 : tunables 0 0 0 : slabdata 1 1 0 blockInfoCache 0 0 4160 7 8 : tunables 0 0 0 : slabdata 0 0 0 fuse_request 0 0 400 10 1 : tunables 0 0 0 : slabdata 0 0 0 fuse_inode 0 0 448 9 1 : tunables 0 0 0 : slabdata 0 0 0 kcopyd_job 0 0 272 15 1 : tunables 0 0 0 : slabdata 0 0 0 dm_uevent 0 0 2464 13 8 : tunables 0 0 0 : slabdata 0 0 0 dm_rq_target_io 0 0 224 18 1 : tunables 0 0 0 : slabdata 0 0 0 ext4_inode_cache 10073 11011 624 13 2 : tunables 0 0 0 : slabdata 847 847 0 ext4_xattr 0 0 48 85 1 : tunables 0 0 0 : slabdata 0 0 0 ext4_free_block_extents 102 102 40 102 1 : tunables 0 0 0 : slabdata 1 1 0 ext4_alloc_context 36 36 112 36 1 : tunables 0 0 0 : slabdata 1 1 0 ext4_prealloc_space 56 56 72 56 1 : tunables 0 0 0 : slabdata 1 1 0 jbd2_journal_handle 170 170 24 170 1 : tunables 0 0 0 : slabdata 1 1 0 jbd2_journal_head 64 64 64 64 1 : tunables 0 0 0 : slabdata 1 1 0 jbd2_revoke_table 256 256 16 256 1 : tunables 0 0 0 : slabdata 1 1 0 jbd2_revoke_record 128 128 32 128 1 : tunables 0 0 0 : slabdata 1 1 0 kmalloc_dma-512 8 8 512 8 1 : tunables 0 0 0 : slabdata 1 1 0 sgpool-128 12 12 2560 12 8 : tunables 0 0 0 : slabdata 1 1 0 sgpool-64 12 12 1280 12 4 : tunables 0 0 0 : slabdata 1 1 0 sgpool-32 12 12 640 12 2 : tunables 0 0 0 : slabdata 1 1 0 RAWv6 11 11 704 11 2 : tunables 0 0 0 : slabdata 1 1 0 UDPLITEv6 0 0 704 11 2 : tunables 0 0 0 : slabdata 0 0 0 UDPv6 11 11 704 11 2 : tunables 0 0 0 : slabdata 1 1 0 tw_sock_TCPv6 0 0 256 16 1 : tunables 0 0 0 : slabdata 0 0 0 TCPv6 12 12 1344 12 4 : tunables 0 0 0 : slabdata 1 1 0 cfq_queue 168 180 112 36 1 : tunables 0 0 0 : slabdata 5 5 0 bsg_cmd 0 0 288 14 1 : tunables 0 0 0 : slabdata 0 0 0 mqueue_inode_cache 14 14 576 14 2 : tunables 0 0 0 : slabdata 1 1 0 hugetlbfs_inode_cache 11 11 360 11 1 : tunables 0 0 0 : slabdata 1 1 0 dquot 0 0 192 21 1 : tunables 0 0 0 : slabdata 0 0 0 posix_timers_cache 0 0 120 34 1 : tunables 0 0 0 : slabdata 0 0 0 UDP-Lite 0 0 576 14 2 : tunables 0 0 0 : slabdata 0 0 0 xfrm_dst_cache 12 12 320 12 1 : tunables 0 0 0 : slabdata 1 1 0 UDP 26 28 576 14 2 : tunables 0 0 0 : slabdata 2 2 0 tw_sock_TCP 21 21 192 21 1 : tunables 0 0 0 : slabdata 1 1 0 TCP 27 39 1216 13 4 : tunables 0 0 0 : slabdata 3 3 0 blkdev_queue 72 72 1328 12 4 : tunables 0 0 0 : slabdata 6 6 0 blkdev_requests 48 60 200 20 1 : tunables 0 0 0 : slabdata 3 3 0 biovec-256 20 20 3072 10 8 : tunables 0 0 0 : slabdata 2 2 0 biovec-128 0 0 1536 10 4 : tunables 0 0 0 : slabdata 0 0 0 biovec-64 0 0 768 10 2 : tunables 0 0 0 : slabdata 0 0 0 bip-256 20 20 3136 10 8 : tunables 0 0 0 : slabdata 2 2 0 bip-128 0 0 1600 10 4 : tunables 0 0 0 : slabdata 0 0 0 bip-64 0 0 832 19 4 : tunables 0 0 0 : slabdata 0 0 0 sock_inode_cache 621 639 448 9 1 : tunables 0 0 0 : slabdata 71 71 0 skbuff_fclone_cache 10 10 384 10 1 : tunables 0 0 0 : slabdata 1 1 0 file_lock_cache 39 39 104 39 1 : tunables 0 0 0 : slabdata 1 1 0 net_namespace 0 0 1184 13 4 : tunables 0 0 0 : slabdata 0 0 0 shmem_inode_cache 1173 1173 456 17 2 : tunables 0 0 0 : slabdata 69 69 0 Acpi-Operand 562 714 40 102 1 : tunables 0 0 0 : slabdata 7 7 0 Acpi-Namespace 782 850 24 170 1 : tunables 0 0 0 : slabdata 5 5 0 task_delay_info 302 306 80 51 1 : tunables 0 0 0 : slabdata 6 6 0 taskstats 24 24 328 12 1 : tunables 0 0 0 : slabdata 2 2 0 proc_inode_cache 509 510 384 10 1 : tunables 0 0 0 : slabdata 51 51 0 sigqueue 29 56 144 28 1 : tunables 0 0 0 : slabdata 2 2 0 radix_tree_node 6189 6617 296 13 1 : tunables 0 0 0 : slabdata 509 509 0 bdev_cache 24 32 512 8 1 : tunables 0 0 0 : slabdata 4 4 0 sysfs_dir_cache 12065 12070 48 85 1 : tunables 0 0 0 : slabdata 142 142 0 inode_cache 1268 1749 360 11 1 : tunables 0 0 0 : slabdata 159 159 0 dentry 24430 24450 136 30 1 : tunables 0 0 0 : slabdata 815 815 0 buffer_head 22332 28544 64 64 1 : tunables 0 0 0 : slabdata 446 446 0 vm_area_struct 14118 14398 88 46 1 : tunables 0 0 0 : slabdata 313 313 0 mm_struct 129 135 448 9 1 : tunables 0 0 0 : slabdata 15 15 0 signal_cache 176 182 576 14 2 : tunables 0 0 0 : slabdata 13 13 0 sighand_cache 174 180 1344 12 4 : tunables 0 0 0 : slabdata 15 15 0 task_struct 292 300 1088 15 4 : tunables 0 0 0 : slabdata 20 20 0 anon_vma 5627 5632 16 256 1 : tunables 0 0 0 : slabdata 22 22 0 idr_layer_cache 543 546 152 26 1 : tunables 0 0 0 : slabdata 21 21 0 kmalloc-8192 4 4 8192 4 8 : tunables 0 0 0 : slabdata 1 1 0 kmalloc-4096 69 72 4096 8 8 : tunables 0 0 0 : slabdata 9 9 0 kmalloc-2048 121 128 2048 8 4 : tunables 0 0 0 : slabdata 16 16 0 kmalloc-1024 224 232 1024 8 2 : tunables 0 0 0 : slabdata 29 29 0 kmalloc-512 1669 1688 512 8 1 : tunables 0 0 0 : slabdata 211 211 0 kmalloc-256 330 448 256 16 1 : tunables 0 0 0 : slabdata 28 28 0 kmalloc-128 7753 8064 128 32 1 : tunables 0 0 0 : slabdata 252 252 0 kmalloc-64 2111 2688 64 64 1 : tunables 0 0 0 : slabdata 42 42 0 kmalloc-32 3179 3200 32 128 1 : tunables 0 0 0 : slabdata 25 25 0 kmalloc-16 3526 3584 16 256 1 : tunables 0 0 0 : slabdata 14 14 0 kmalloc-8 5629 5632 8 512 1 : tunables 0 0 0 : slabdata 11 11 0 kmalloc-192 312 399 192 21 1 : tunables 0 0 0 : slabdata 19 19 0 kmalloc-96 1082 1218 96 42 1 : tunables 0 0 0 : slabdata 29 29 0
vmstate -mjustin@Justin:~$ vmstat -m Cache Num Total Size Pages VMBlockInodeCache 7 7 4480 7 blockInfoCache 0 0 4160 7 fuse_request 0 0 400 10 fuse_inode 0 0 448 9 kcopyd_job 0 0 272 15 dm_uevent 0 0 2464 13 dm_rq_target_io 0 0 224 18 ext4_inode_cache 9945 10907 624 13 ext4_xattr 0 0 48 85 ext4_free_block_extents 102 102 40 102 ext4_alloc_context 36 36 112 36 ext4_prealloc_space 56 56 72 56 jbd2_journal_handle 170 170 24 170 jbd2_journal_head 64 64 64 64 jbd2_revoke_table 256 256 16 256 jbd2_revoke_record 128 128 32 128 kmalloc_dma-512 8 8 512 8 sgpool-128 12 12 2560 12 sgpool-64 12 12 1280 12 sgpool-32 12 12 640 12 RAWv6 11 11 704 11 Cache Num Total Size Pages UDPLITEv6 0 0 704 11 UDPv6 11 11 704 11 tw_sock_TCPv6 0 0 256 16 TCPv6 12 12 1344 12 cfq_queue 168 180 112 36 bsg_cmd 0 0 288 14 mqueue_inode_cache 14 14 576 14 hugetlbfs_inode_cache 11 11 360 11 dquot 0 0 192 21 posix_timers_cache 0 0 120 34 UDP-Lite 0 0 576 14 xfrm_dst_cache 12 12 320 12 UDP 26 28 576 14 tw_sock_TCP 21 21 192 21 TCP 27 39 1216 13 blkdev_queue 72 72 1328 12 blkdev_requests 48 60 200 20 biovec-256 20 20 3072 10 biovec-128 0 0 1536 10 biovec-64 0 0 768 10 bip-256 20 20 3136 10 Cache Num Total Size Pages bip-128 0 0 1600 10 bip-64 0 0 832 19 sock_inode_cache 623 639 448 9 skbuff_fclone_cache 10 10 384 10 file_lock_cache 39 39 104 39 net_namespace 0 0 1184 13 shmem_inode_cache 1173 1173 456 17 Acpi-Operand 562 714 40 102 Acpi-Namespace 782 850 24 170 task_delay_info 302 306 80 51 taskstats 24 24 328 12 proc_inode_cache 502 520 384 10 sigqueue 29 56 144 28 radix_tree_node 5994 6474 296 13 bdev_cache 24 32 512 8 sysfs_dir_cache 12065 12070 48 85 inode_cache 1254 1694 360 11 dentry 23785 24480 136 30 buffer_head 22158 28544 64 64 vm_area_struct 14176 14444 88 46 mm_struct 129 135 448 9 Cache Num Total Size Pages signal_cache 177 182 576 14 sighand_cache 175 180 1344 12 task_struct 292 300 1088 15 anon_vma 5627 5632 16 256 idr_layer_cache 543 546 152 26 kmalloc-8192 4 4 8192 4 kmalloc-4096 70 72 4096 8 kmalloc-2048 125 128 2048 8 kmalloc-1024 225 232 1024 8 kmalloc-512 1670 1688 512 8 kmalloc-256 336 448 256 16 kmalloc-128 7823 8064 128 32 kmalloc-64 2091 2688 64 64 kmalloc-32 3179 3200 32 128 kmalloc-16 3526 3584 16 256 kmalloc-8 5629 5632 8 512 kmalloc-192 305 399 192 21 kmalloc-96 1082 1218 96 42
vmstat [-S unit]vmstat [-S unit]
The -S followed by k or K or m or M switches changes the units of ouput from bytes to outputs between 1000, 1024, 1000000, or 1048576 bytes.
Note this does not change the swap (si/so) or block (bi/bo) fields.
NOTE: default is -S K
vmstat [-p partition]FIELD DESCRIPTION FOR DISK PARTITION MODE
reads: Total number of reads issued to this partition read sectors: Total read sectors for partition writes : Total number of writes issued to this partition requested writes: Total number of write requests made for partition
The -p followed by some partition name for detailed statistics (2.5.70 or above required)
justin@Justin:~$ vmstat -p /dev/sda3
sda3 reads read sectors writes requested writes
74214 1862216 21549 497241
vmstat [-d] / vmstat [-D]donot repeatThe -d reports disk statistics (2.5.70 or above required)
FIELD DESCRIPTION FOR DISK MODE
Reads total: Total reads completed successfully merged: grouped reads (resulting in one I/O) sectors: Sectors read successfully ms: milliseconds spent reading
Writes total: Total writes completed successfully merged: grouped writes (resulting in one I/O) sectors: Sectors written successfully ms: milliseconds spent writing
IO cur: I/O in progress s: seconds spent for I/O
justin@Justin:~$ vmstat -d
disk- ------------reads------------ ------------writes----------- -----IO------
total merged sectors ms total merged sectors ms cur sec
fd0 4 0 32 128 0 0 0 0 0 0
sr0 0 0 0 0 0 0 0 0 0 0
sda 96046 84200 2917603 1110920 49705 222158 2174193 7057256 0 745
sdb 18642 465040 691231 79184 4 14 25 16 0 47
sdc 15601 16090 240327 56472 3 7 17 20 0 37
sdd 21723 29590 399857 158836 3 7 17 16 0 107
sde 87940 42772 2586593 1581932 7159 19654 196392 256476 0 662
dm-0 127560 0 2580698 2931840 24549 0 196392 1152800 0 661
loop0 0 0 0 0 0 0 0 0 0 0
loop1 0 0 0 0 0 0 0 0 0 0
loop2 0 0 0 0 0 0 0 0 0 0
loop3 0 0 0 0 0 0 0 0 0 0
loop4 0 0 0 0 0 0 0 0 0 0
loop5 0 0 0 0 0 0 0 0 0 0
loop6 0 0 0 0 0 0 0 0 0 0
loop7 0 0 0 0 0 0 0 0 0 0
loop8 0 0 0 0 0 0 0 0 0 0
loop9 0 0 0 0 0 0 0 0 0 0
loop10 0 0 0 0 0 0 0 0 0 0
loop11 0 0 0 0 0 0 0 0 0 0
loop12 0 0 0 0 0 0 0 0 0 0
disk- ------------reads------------ ------------writes----------- -----IO------
The -D reports some summary statistics about disk activity.
justin@Justin:~$ vmstat -D
72 disks
7 partitions
367882 total reads
637751 merged reads
9430861 read sectors
5927908 milli reading
81866 writes
242220 merged writes
2573500 written sectors
8473444 milli writing
0 inprogress IO
2265 milli spent IO
DisplayProcsProcs r: The number of processes waiting for run time. b: The number of processes in uninterruptible sleep.
b: 处在非中断睡眠状态的进程数 (在等待io的进程数 ???)
w 在Linux版本上的vmstat没有：（AIX..)
w: 被交换出去的可运行的进程数(可以进入运行队列但被替换的进程)。此数由 linux 计算得出，但 linux 并不耗尽交换空间
A kernel thread (or a normal process) can be in any of the following process states:
- running
- interruptible
- uninterruptible
- zombie
- stopped
- traced
- dead
These states are defined in include/linux/sched.h:
A process in the running state (TASK_RUNNING) is in the scheduler run queue and is a candidate for getting CPU time allotted by the scheduler.
A task in the interruptible state (TASK_INTERRUPTIBLE) is waiting for an event to occur and is not in the scheduler run queue. When the task gets woken up, or if a signal is delivered to it, it re-enters the run queue.
The uninterruptible state (TASK_UNINTERRUPTIBLE) is similar to the interruptible state except that receipt of a signal will not put the task back into the run queue.
NOTE: 睡眠，不可中断（不可被信号唤醒）
与TASK_INTERRUPTINLE状态的唯一区别就是该状态不可被收到的信号唤醒。这种状态很少用到，但在一些特殊的情况下（进程必须等待，直到一个不能被中断的事件发生，发送硬盘I/O要求而等待I/O完成的状态，等待TTY 终端的输入的状态等），这种状态是很有用的。例如，当进程打开一个设备文件，其相应的设备驱动程序开始探测相应的硬件设备时会用到这种状态。探测完成以前，设备驱动程序不能被中断，否则，硬件设备会处于不可预知的状态。在状态通常在进程需要不受干扰的等待或者所等待的事件会很快发生时使用。
A stopped task (TASK_STOPPED) has stopped execution due to receipt of certain signals. (Ctrl+Z)
If an application such as strace is using the ptrace support in the kernel to intercept a task, it'll be in the traced state (TASK_TRACED).
A task in the zombie state (EXIT_ZOMBIE) has terminated, but its parent did not wait for the task to complete.
An exiting task is either in the EXIT_ZOMBIE state or the dead (EXIT_DEAD) state.
Metrics:r+b = cpu load < num of cpu
一般来说，cpu load长期大于cpu个数，说明cpu负载高，会影响系统性能
MemoryMemory 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.
inact: the amount of inactive memory. (-a option) active: the amount of active memory. (-a option)
Memory swpd: 虚拟内存使用情况，单位：KB free: 空闲的内存，单位KB buff: 被用来做为缓存的内存数，单位：KB
Buffers and Cache (Too Much Physical Memory) ===============================================
Cache:缓存 if your system has much more physical memory than required by your applications, Linux will cache recently used files in physical memory so that subsequent accesses to that file do not require an access to the hard drive. This can greatly speed up applications that access the hard drive frequently, which, obviously, can prove especially useful for frequently launched applications.
Note : most tools that report statistics about “cache” are actually referring to disk cache.
**Buffer:**缓冲
In addition to cache, Linux also uses extra memory as buffers. To further optimize applications, Linux sets aside memory to use for data that needs to be written to disk. These set-asides are called buffers. If an application has to write something to the disk, which would usually take a long time, Linux lets the application continue immediately but saves the file data into a memory buffer. At some point in the future, the buffer is flushed to disk, but the application can continue immediately.
Metrics:Swapsi: Amount of memory swapped in from disk (/s). so: Amount of memory swapped to disk (/s).
Swap
si: 从磁盘交换到内存的交换页数量，单位：KB/秒? so: 从内存交换到磁盘的交换页数量，单位：KB/秒?
unit not in pages/s, but in KB/s
getconf PAGESIZE 4096
Metrics:si, so 的值一般都为0, 如果长期不为0,表示系统内存不足， 需要增加系统内存。
IOblocks differs between different devicesbi: Blocks received from a block device (blocks/s). bo: Blocks sent to a block device (blocks/s).
oldlinux － 512B 1KB ...
NOTES:
All linux blocks are currently 1024 bytes. Old kernels may report blocks as 512 bytes, 2048 bytes, or 4096 bytes.
Metrics:bi+bo 的参考值为1000,如果超过1000, 说明块设备读写频繁，
如果这个时候cpu的wa时间较大，表示系统磁盘I/O有问题，应该考虑提高磁盘的读写性能。
SystemSystem
in: 每秒的中断数，包括时钟中断 (interrupt) cs: 每秒的环境（上下文）切换次数 (context switch)
System in: The number of interrupts per second, including the clock. cs: The number of context switches per second.
Metrics:这两个值越大，标明系统消耗在进程上下文切换中的cpu时间越多。
CPUCPU These are percentages of total CPU time. 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.
justin@Justin:~$ vmstat procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa 0 0 0 22024 34488 510808 0 0 1305 45 342 1460 22 18 35 25
其中:
如果 r 经常大于 4 ,且id经常少于40，表示cpu的负荷很重。
b 经常大于3,wa>30% 表示 io性能不好 (io性能可用iostat确认)
for (every per-processor cpu_stat structure) do
if (current cpu is quiesced) increment idle; else if (current cpu is on the interrupt stack) increment kernel; else if (current cpu is running this clock handler code) increment kernel; else if (current cpu is running the idle thread) if (wait_value from first loop is non-zero) increment waitio; else (wait_value from first loop was zero) increment idle; else if (current cpu lwp state is USER) increment user; done
"I/O Wait" is a measurement of CPU idle time. It's a common misconception that "I/O Wait" cycles are blocked, and can not be used by other processes. This is not the case.
This is illustrated in the example from John Moore (below). The example starts with an idle system, then adds an I/O intensive workload (WIO = 87%). Next a CPU intensive application is added. The CPU intensive applications uses all of the "Wait I/O" cycles (WIO=0%), but does not change the I/O rate.
The practical implications of "Wait I/O" not being blocked include:
Adding CPU's will not reduce Wait I/O. CPU statistics should be measured as Utilization = usr + sys Idle = wait + idle High CPU utilization can mask I/O bottlenecks You can mix I/O intensive with CPU intensive applications on same server and not significantly affect performance.
Metrics:us+sy 是系统运行程序消耗的cpu比例，如果>80,说明系统cpu资源不足：
id＋wa是cpu的空闲时间，最好在40%左右，
wa是系统cpu等待io的时间，wa超过20%，说明io等待严重，可能由于 大量磁盘随机读写造成，或者磁盘、磁盘控制器的带宽瓶颈。（相对瓶颈）
这个是相对瓶颈，但是如果系统cpu负载很大，则也能引起wa降低，相对来说io等待就不是瓶颈了。
NOTESWiki comments powered by DisqusSince procps 3.1.9, vmstat lets you choose units (k, K, m, M) default is K (1024 bytes) in the default mode
Contact: mmmyddd@github 2009-2019 © CopyLeft Contributed |
[ Top | Up | Dir | Home ] |
|||||
