vmstat Command 参考

vmstat Command

Purpose

       Reports virtual memory statistics.

Syntax

       vmstat [ -f ] [ -i ] [ -s  ] [ -I ] [ -t ] [ -v ] [ -w] [ -l ] [ { -p | -P } pagesize | ALL ] ALL ] [ PhysicalVolume ... ] [ Interval [ Count ] ]

Description

       The vmstat command reports statistics about kernel threads, virtual memory, disks, traps and pocessor 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. If the Interval parameter is not specified, the vmstat command
       generates a single report that contains statistics for the time since system startup 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 processor 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 processor was busy in usr mode
       at the time of the clock interrupt, then usr gets the clock tick added into its category. If the processor was busy in kernel mode at the time of
       the clock interrupt, then the sys category gets the tick. If the processor 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 processor is not busy, the idle category gets
       the tick. The inflated view of wio time results from all idle processors 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 processors it has. The
       wio time is reported by the commands sar (%wio), vmstat (wa) and iostat (% iowait).

       With AIX 5.4 and later versions, you can use the accounting system based on the Scaled Performance Utilization Resources Register (SPURR). The SPURR
       is supported on the POWER6 processors. It is similar to the Performance Utilization Resources Register (PURR), except that it scales as a function
       of degree of processor throttling. If the hardware supports the SPURR, the processor utilization statistics shown by the vmstat command is
       proportional to the frequency or the instruction dispatch rate of the processor. The processor utilization statistics are capped to the PURR values
       in the turbo mode if the turbo mode accounting is disabled. You can enable the turbo mode accounting system through the System Management Interface
       Tool (SMIT).

       The kernel maintains statistics for kernel threads, paging, and interrupt activity, which the vmstat command accesses through the use of the the
       perfstat kernel extension. 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.

       Beginning with AIX 5.3, the vmstat command reports the number of physical processors consumed (pc), and the percentage of entitlement consumed (ec),
       in Micro-Partitioning environments. These metrics will only be displayed on Micro-Partitioning environments.

       Reports generated by the vmstat command contains the following column headings and their description:

       kthr: information about kernel thread states.
       r

            Average number of runnable kernel threads over the sampling interval. Runnable refers to threads that are ready but waiting to run and to those
            threads already running.
       b
            Average number of kernel threads placed in the VMM wait queue (awaiting resource, awaiting input/output) over the sampling interval.

       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 processor time.
       us
            User time.
       sy
            System time.
       id
            Processor idle time.
       wa
            Processor idle time during which the system had outstanding disk/NFS I/O request(s). See detailed description above.
       pc
            Number of physical processors consumed. Displayed only if the partition is running with shared processor.
       ec
            The percentage of entitled capacity consumed. Displayed only if the partition is running with shared processor. Because the time base over
            which this data is computed can vary, the entitled capacity percentage can sometimes exceed 100%. This excess is noticeable only with small
            sampling intervals.

       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 PhysicalVolume parameter is used, the physical volume names are printed at the beginning of command execution.

       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 I/O to raw devices 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.
       If, while the vmstat command is running, there is a change in system configuration that will affect the output, vmstat prints a warning message
       about the configuration change. It then continues the output, after printing the updated system configuration information and the header.

       If the -l flag is specified, an additional "large-page" section is displayed with the following columns:
       alp
            Indicates the number of large pages currently in use.
       flp
            Indicates the number of large pages on the large page freelist.

       If the -p option is specified, additional lines of VMM statistics are displayed for the specified page sizes. With -I and -t options, the -p option
       produces an additional line for the specified page size. This line contains the following VMM statistics relevant to the specified page size:
       *    avm
       *    fre
       *    re
       *    fi
       *    fo
       *    pi
       *    po
       *    fr
       *    sr
       *    cy Note: The disply of the re, fi, fo, and cy options are affected by the -I option.

       These VMM statistics are preceded by a psz column and followed by an siz column. The description of these two columns follows:
       psz
            Page size (for example, 4K, 64K).
       siz
            Number of frames of the specified page size that exist on the system.

       With the -s option, the -p option produces a separate stanza of output that contains only the statistics relevant to the specified page size. This
       additional stanza is preceded by a page size header.

       The -P option produces the following report for the specified page size:
       pgsz
            Indicates the page size (for example, 4K, 64K).
       Memory
            Indicates the memory statistics for the specified page sizes.
       siz
            The number of frames of the specified page size that exist on the system.
       avm
            Active virtual pages applicable to the specified page size.
       fre
            Size of the free list for the specified page size.
       Page
            Indicates the relevant page faults and paging activity for the specified page size. The page related columns re, pi, po, fr, sr, cy, fi, and fo
            are also applicable to this report.

Flags
       Note: If the -f (or -s) flag is entered on the command line, then the system will only accept the -f (or -s) flag and will ignore other flags. If
       both -f and -s flags are specified, the system will accept only the first flag and ignore 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. Note: The -I, -t, -w, and -l flags are ignored when they are
            specified with the -i flag.
       -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.
       -l
            Displays an additional "large-page" section with the alp and flp columns.
       -p pagesize
            Appends the VMM statistics for the specified page size to the regular vmstat output.
       -P pagesize
            Displays only the VMM statistics which are relevant for the specified page size.
       -s
            Writes to standard output the contents of the sum structure, which contains an absolute count of paging events since system initialization. The
            -s flag can only be used with the -v flag. 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.)
              free frame waits
                   Incremented each time a process requests a page frame, the free list is empty, and the process is forced to wait while the free list is
                   replenished.
              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 processor 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.
              decrementer interrupts
                   Incremented on each decrementer interrupt.
              mpc send interrupts
                   Incremented on each mpc send interrupt
              mpc receive interrupts
                   Incremented on each mpc receive interrupt
              phantom interrupts
                   Incremented on each phantom interrupt
              traps
                   Not maintained by the operating system.
              syscalls
                   Incremented for each system call.
            When used with the -p pagesize option, the -s option appends the sum structure for the specified page size to the system-wide sum structure.
            This additional stanza is preceded by a page size header (for example, 4K pages). The following details are not be displayed in this pagesize-
            based stanza as these statistics are not related to page sizes:
              *    Processor context switches
              *    Device interrupts
              *    Software interrupts
              *    Decrementer interrupts
              *    MPC-sent interrupts
              *    MPC-received interrupts
              *    Phantom interrupts
              *    Traps
              *    Syscalls
       -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.
       -v
            Writes to standard output various statistics maintained by the Virtual Memory Manager. The -v flag can only be used with the -s flag.
              memory pages
                   Size of real memory in number of 4 KB pages.
              lruable pages
                   Number of 4 KB pages considered for replacement. This number excludes the pages used for VMM internal pages, and the pages used for the
                   pinned part of the kernel text.
              free pages
                   Number of free 4 KB pages.
              memory pools
                   Tuning parameter (managed using vmo) specifying the number of memory pools.
              pinned pages
                   Number of pinned 4 KB pages.
              maxpin percentage
                   Tuning parameter (managed using vmo) specifying the percentage of real memory which can be pinned.
              minperm percentage
                   Tuning parameter (managed using vmo) in percentage of real memory. This specifies the point below which file pages are protected from
                   the re-page algorithm.
              maxperm percentage
                   Tuning parameter (managed using vmo) in percentage of real memory. This specifies the point above which the page stealing algorithm
                   steals only file pages.

              numperm percentage
                   Percentage of memory currently used by the file cache.
              file pages
                   Number of 4 KB pages currently used by the file cache.
              compressed percentage
                   Percentage of memory used by compressed pages.
              compressed pages
                   Number of compressed memory pages.
              numclient percentage
                   Percentage of memory occupied by client pages.
              maxclient percentage
                   Tuning parameter (managed using vmo) specifying the maximum percentage of memory which can be used for client pages.
              client pages
                   Number of client pages.
              remote pageouts scheduled
                   Number of pageouts scheduled for client filesystems.
              pending disk I/Os blocked with no pbuf
                   Number of pending disk I/O requests blocked because no pbuf was available. Pbufs are pinned memory buffers used to hold I/O requests at
                   the logical volume manager layer.
              paging space I/Os blocked with no psbuf
                   Number of paging space I/O requests blocked because no psbuf was available. Psbufs are pinned memory buffers used to hold I/O requests
                   at the virtual memory manager layer.
       -v
            (Statistics displayed by -v, continued):
              filesystem I/Os blocked with no fsbuf
                   Number of filesystem I/O requests blocked because no fsbuf was available. Fsbuf are pinned memory buffers used to hold I/O requests in
                   the filesystem layer.
              client filesystem I/Os blocked with no fsbuf
                   Number of client filesystem I/O requests blocked because no fsbuf was available. NFS (Network File System) and VxFS (Veritas) are client
                   filesystems. Fsbuf are pinned memory buffers used to hold I/O requests in the filesystem layer.
              external pager filesystem I/Os blocked with no fsbuf
                   Number of external pager client filesystem I/O requests blocked because no fsbuf was available. JFS2 is an external pager client
                   filesystem. Fsbuf are pinned memory buffers used to hold I/O requests in the filesystem layer.
              Virtualized Partition Memory Page Faults
                   Total number of virtual partition memory page faults recorded for the virtualized partition.
              Time resolving virtualized partition memory page faults
                   Total time (in seconds, with millisecond granularity) the virtual partition has been blocked waiting for the resolution of the virtual
                   partition's memory page faults.
       -w
            Display the report in wide mode

Examples
       1    To display a summary of the statistics since boot, enter:

            vmstat
       2    To display five summaries at 2-second intervals, enter:

            vmstat 2 5
       3    To display a summary of the statistics since boot including statistics for logical disks scdisk13 and scdisk14, enter:

            vmstat scdisk13 scdisk14
       4    To display fork statistics, enter:

            vmstat  -f
       5    To display the count of various events, enter:

            vmstat -s
       6    To display time-stamp next to each column of output of vmstat, enter:

            vmstat -t
       7    To display the I/O oriented view with an alternative set of columns, enter:

            vmstat -I
       8    To display all the VMM statistics available, enter:

            vmstat -vs
       9    To display the large-page section with the alp and flp columns at 8-second intervals, enter:

            vmstat -l 8
       10   To display the VMM statistics specific to a particular page size (in the example, 4 K), enter:

            vmstat -p 4 K
       11   To display the VMM statistics for all page sizes that are supported on the system, enter:

            vmstat -p ALL
            OR

            vmstat -p all
       12   To display only the VMM statistics for a particular page size (in this example, 4K), enter:

            vmstat -P 4K
       13   To display only the per-page breakdown of VMM statistics for all supported page sizes, enter:

            vmstat -P ALL
            OR

            vmstat -P all

Files

       /usr/bin/vmstat
            Contains the vmstat command.

Related Information

       The iostat and vmo command.

       Memory performance in Performance management.