Commands Reference, Volume 6

vmstat Command

Syntax

vmstat [ -f ] [ -i ] [ -s ] [ -I ] [ -t ] [ PhysicalVolume ... ] [ Interval [ Count ] ]

The vmstat command reports statistics about kernel threads, virtual memory, disks, traps and CPU activity. Reports generated by the vmstat command can be used to balance system load activity. These system-wide statistics (among all processors) are calculated as averages for values expressed as percentages, and as sums otherwise.

If the vmstat command is invoked without flags, the report contains a summary of the virtual memory activity since system startup. If the -f flag is specified, the vmstat command reports the number of forks since system startup. The PhysicalVolume parameter specifies the name of the physical volume.

The Interval parameter specifies the amount of time in seconds between each report. The first report contains statistics for the time since system startup. Subsequent reports contain statistics collected during the interval since the previous report. If the Interval parameter is not specified, the vmstat command generates a single report and then exits. The Count parameter can only be specified with the Interval parameter. If the Count parameter is specified, its value determines the number of reports generated and the number of seconds apart. If the Interval parameter is specified without the Count parameter, reports are continuously generated. A Count parameter of 0 is not allowed.

AIX 4.3.3 and later contain enhancements to the method used to compute the percentage of CPU time spent waiting on disk I/O (wio time).The method used in AIX 4.3.2 and earlier versions of the operating system can, under certain circumstances, give an inflated view of wio time on SMPs.

The method used in AIX 4.3.2 and earlier versions is as follows: At each clock interrupt on each processor (100 times a second per processor), a determination is made as to which of the four categories (usr/sys/wio/idle) to place the last 10 ms of time. If the CPU was busy in usr mode at the time of the clock interrupt, then usr gets the clock tick added into its category. If the CPU was busy in kernel mode at the time of the clock interrupt, then the sys category gets the tick. If the CPU was not busy, a check is made to see if any I/O to disk is in progress. If any disk I/O is in progress, the wio category is incremented. If no disk I/O is in progress and the CPU is not busy, the idle category gets the tick. The inflated view of wio time results from all idle CPUs being categorized as wio regardless of the number of threads waiting on I/O. For example, systems with just one thread doing I/O could report over 90 percent wio time regardless of the number of CPUs it has. The wio time is reported by the commands sar (%wio), vmstat (wa) and iostat (% iowait).

The method used in operating system AIX 4.3.3 and later is as follows: The change in operating system AIX 4.3.3 is to only mark an idle CPU as wio if an outstanding I/O was started on that CPU. This method can report much lower wio times when just a few threads are doing I/O and the system is otherwise idle. For example, a system with four CPUs and one thread doing I/O will report a maximum of 25 percent wio time. A system with 12 CPUs and one thread doing I/O will report a maximum of 8 percent wio time. NFS client reads/writes go through the VMM, and the time that biods spend in the VMM waiting for an I/O to complete is now reported as I/O wait time.

The kernel maintains statistics for kernel threads, paging, and interrupt activity, which the vmstat command accesses through the use of the knlist subroutine and the /dev/kmem pseudo-device driver. The disk input/output statistics are maintained by device drivers. For disks, the average transfer rate is determined by using the active time and number of transfers information. The percent active time is computed from the amount of time the drive is busy during the report.

The following example of a report generated by the vmstat command contains the column headings and their description:

kthr: kernel thread state changes per second over the sampling interval.

`r`	Number of kernel threads placed in run queue.
`b`	Number of kernel threads placed in wait queue (awaiting resource, awaiting input/output).

Memory: information about the usage of virtual and real memory. Virtual pages are considered active if they have been accessed. A page is 4096 bytes.

avm Active virtual pages.

fre

Size of the free list.

Note: A large portion of real memory is utilized as a cache for file system data. It is not unusual for the size of the free list to remain small.

Page: information about page faults and paging activity. These are averaged over the interval and given in units per second.

`re`	Pager input/output list.
`pi`	Pages paged in from paging space.
`po`	Pages paged out to paging space.
`fr`	Pages freed (page replacement).
`sr`	Pages scanned by page-replacement algorithm.
`cy`	Clock cycles by page-replacement algorithm.

Faults: trap and interrupt rate averages per second over the sampling interval.

`in`	Device interrupts.
`sy`	System calls.
`cs`	Kernel thread context switches.

Cpu: breakdown of percentage usage of CPU time.

`us`	User time.
`sy`	System time.
`id`	CPU idle time.
`wa`	CPU idle time during which the system had outstanding disk/NFS I/O request(s). See detailed description above.

Disk: Provides the number of transfers per second to the specified physical volumes that occurred in the sample interval. The PhysicalVolume parameter can be used to specify one to four names. Transfer statistics are given for each specified drive in the order specified. This count represents requests to the physical device. It does not imply an amount of data that was read or written. Several logical requests can be combined into one physical request.

If the -I flag is specified, an I/O oriented view is presented with the following column changes.

`kthr`	The column p will also be displayed besides columns r and b. p Number of threads waiting on actual physical I/O per second.
`page`	New columns fi and fo will be displayed instead of re and cy columns. fi File page-ins per second. fo File page-outs per second.

Flags

Note: Both the -f and -s flags can be entered on the command line, but the system will only accept the first flag specified and override the second flag.

-f	Reports the number of forks since system startup.
-i	Displays the number of interrupts taken by each device since system startup.
-I	Displays I/O oriented view with the new columns of output, p under heading `kthr`,and columns fi and fo under heading page instead of the columns re and cy in the page heading.
-s	Writes to standard output the contents of the sum structure, which contains an absolute count of paging events since system initialization. The -s option is exclusive of the other vmstat command options. These events are described as follows: address translation faults Incremented for each occurrence of an address translation page fault. I/O may or may not be required to resolve the page fault. Storage protection page faults (lock misses) are not included in this count. page ins Incremented for each page read in by the virtual memory manager. The count is incremented for page ins from page space and file space. Along with the page out statistic, this represents the total amount of real I/O initiated by the virtual memory manager. page outs Incremented for each page written out by the virtual memory manager. The count is incremented for page outs to page space and for page outs to file space. Along with the page in statistic, this represents the total amount of real I/O initiated by the virtual memory manager. paging space page ins Incremented for VMM initiated page ins from paging space only. paging space page outs Incremented for VMM initiated page outs to paging space only. total reclaims Incremented when an address translation fault can be satisfied without initiating a new I/O request. This can occur if the page has been previously requested by VMM, but the I/O has not yet completed; or if the page was pre-fetched by VMM's read-ahead algorithm, but was hidden from the faulting segment; or if the page has been put on the free list and has not yet been reused. zero-filled page faults Incremented if the page fault is to working storage and can be satisfied by assigning a frame and zero-filling it. executable-filled page faults Incremented for each instruction page fault. pages examined by the clock VMM uses a clock-algorithm to implement a pseudo least recently used (lru) page replacement scheme. Pages are aged by being examined by the clock. This count is incremented for each page examined by the clock. revolutions of the clock hand Incremented for each VMM clock revolution (that is, after each complete scan of memory). pages freed by the clock Incremented for each page the clock algorithm selects to free from real memory.
	backtracks Incremented for each page fault that occurs while resolving a previous page fault. (The new page fault must be resolved first and then initial page faults can be backtracked.) lock misses VMM enforces locks for concurrency by removing addressability to a page. A page fault can occur due to a lock miss, and this count is incremented for each such occurrence. free frame waits Incremented each time a process is waited by VMM while free frames are gathered. extend XPT waits Incremented each time a process is waited by VMM due to a commit in progress for the segment being accessed. pending I/O waits Incremented each time a process is waited by VMM for a page-in I/O to complete. start I/Os Incremented for each read or write I/O request initiated by VMM. This count should equal the sum of page-ins and page-outs. iodones Incremented at the completion of each VMM I/O request. CPU context switches Incremented for each CPU context switch (dispatch of a new process). device interrupts Incremented on each hardware interrupt. software interrupts Incremented on each software interrupt. A software interrupt is a machine instruction similar to a hardware interrupt that saves some state and branches to a service routine. System calls are implemented with software interrupt instructions that branch to the system call handler routine. traps Not maintained by the operating system. syscalls Incremented for each system call.
-t	Prints the time-stamp next to each line of output of vmstat. The time-stamp is displayed in the `HH:MM:SS` format. Note: Time stamp will not be printed if -f, -s, or -i flags are specified.

Examples

To display a summary of the statistics since boot, enter:
```
vmstat
```
To display five summaries at 2-second intervals, enter:
```
vmstat 2 5
```
The first summary contains statistics for the time since boot.
To display a summary of the statistics since boot including statistics for logical disks scdisk13 and scdisk14, enter:
```
vmstat scdisk13 scdisk14
```
To display fork statistics, enter:
vmstat -f
To display the count of various events, enter:
```
vmstat -s
```
To display time-stamp next to each column of output of vmstat, enter:
```
vmstat -t
```
To display the new I/O oriented view with an alternative set of columns, enter:
```
vmstat -I
```

Files

/unix

Symbolic link to the kernel boot image.

/usr/bin/vmstat

Contains the vmstat command.

Related Information

The iostat command.

The knlist subroutine.

The /dev/kmem special file.

Monitoring and Tuning Memory Use in AIX 5L Version 5.1 Performance Management Guide.