The Virtual Memory Manager (VMM) maintains a list of free real-memory page
frames. These page frames are available to hold virtual-memory pages
needed to satisfy a page fault. When the number of pages on the free
list falls below that specified by the MinFree parameter, the VMM
begins to steal pages to add to the free list. The VMM continues to
steal pages until the free list has at least the number of pages specified by
the MaxFree parameter.
If the number of file pages (permanent pages) in memory is less than the
number specified by the minperm parameter, the VMM steals frames
from either computational or file pages, regardless of repage rates. If
the number of file pages is greater than the number specified by the
MaxPerm parameter, the VMM steals frames only from file
pages. Between the two, the VMM normally steals only file pages, but if
the repage rate for file pages is higher than the repage rate for
computational pages, computational pages are stolen as well.
If a process appears to be reading sequentially from a file, the values
specified by the MinPgAhead parameter determine the number of pages
to be read ahead when the condition is first detected. The value
specified by the MaxPgAhead parameter sets the maximum number of
pages that will be read ahead, regardless of the number of preceding
sequential reads.
AIX Version 4 of the operating system allows tuning of the number of file
system bufstructs (Numfsbuf) and the amount of data
processed by the write-behind algorithm (Nmclust).
In AIX Version 4 of the operating system, you can also modify the
thresholds that are used to decide when the system is running out of paging
space. The NpsWarn parameter specifies the number of
paging-space pages available at which the system begins warning processes that
paging space is low. The NpsKill parameter specifies the
number of paging-space pages available at which the system begins killing
processes to release paging space.
| -a
| Displays the current statistic counters.
|
| -b Numfsbuf
| Specifies the number of file system bufstructs. The
default value is 64.
|
| -B Numpbuf
| Specifies the number of pbufs used by the LVM. The maximum value
is 128.
|
| -c Nmclust
| Specifies the number of 16KB clusters processed by write behind.
The default value is 1.
|
| -C
| Enables page coloring for specific hardware platforms. When
enabled, real memory is carefully assigned to virtual memory. On a
system with a direct-mapped cache and certain workloads, this can provide more
constant system performance. The bosboot command must be run
and the system rebooted for this change to take affect.
|
| -d
| Turns on or off deferred paging space allocation. By default, disk
blocks for paging space are not allocated until pageout is actually
required. This option allows this behavior to change so that the disk
block is allocated when the memory is referenced.
|
| -f MinFree
| Specifies the minimum number of frames on the free list. This
number can range from 8 to 204800.
|
| -F MaxFree
| Specifies the number of frames on the free list at which page stealing is
to stop. This number can range from 16 to 204800 but must be greater
than the number specified by the MinFree parameter by at least the
value of MaxPgAhead.
|
| -g LargePageSize
| Specifies the size in bytes of the hardware-supported large pages used
for the implementation for the shmget( ) system call with the
SHM_LGPAGE flag. This must be enabled with a non-zero value
for the -L flag and the bosboot command must be run and
the system rebooted for this change to take affect.
|
| -h
| Specifies that maxperm (-P) should be hard limit. By
default it is a soft limit and numperm is allowed to grow beyond
maxperm as long as there is free real memory available.
|
| -k NpsKill
| Specifies the number of free paging-space pages at which AIX begins
killing processes.
|
| -l LruBucket
| Specifies the size (in 4K pages) of the least recently used (lru)
page-replacement bucket size. This is the number of page frames which
will be examined at one time for possible pageouts when a free frame is
needed. A lower number will result in lower latency when looking for a
free frame, but will also result in behavior that is not as much like a true
lru algorithm. The default value is 512MB and the minimum is
256MB. Tuning this option is not recommended.
|
| -L LargePages
| Specifies the number of large pages to reserve for implementing the
shmget( ) system call with the SHM_LGPAGE flag.
For this change to take effect, you must specify the -g flag, run
the bosboot command, and reboot the system.
|
| -M MaxPin
| Specifies the maximum percentage of real memory that can be
pinned. The default value is 80. If this value is changed, the
new value should ensure that at least 4MB of real memory will be left unpinned
for use by the kernel.
|
| -n uid
| Specifies that processes with uid less than uid should not be
killed when paging space is low. Setting this to 1 would
prevent root processes from being killed.
|
| -N Pd_Npages
| Specifies the number of pages that should be deleted in one chunk from
RAM when a file is deleted. The default value is the largest possible
file size divided by the page size (currently 4096). If the largest
possible file size is 2GB, then Pd_Npages is by default
524288. Tuning this option is really only useful for real-time
applications.
|
| -p minperm
| Specifies the point below which file pages are protected from the repage
algorithm. This value is a percentage of the total real-memory page
frames in the system. The specified value must be greater than or equal
to 5.
|
| -P MaxPerm
| Specifies the point above which the page stealing algorithm steals only
file pages. This value is expressed as a percentage of the total
real-memory page frames in the system. The specified value must be
greater than or equal to 5.
|
| -r MinPgAhead
| Specifies the number of pages with which sequential read-ahead
starts. This value can range from 0 through 4096. It should be a
power of 2.
|
| -R
MaxPgAhead
| Specifies the maximum number of pages to be read ahead. This value
can range from 0 through 4096. It should be a power of 2 and should be
greater than or equal to MinPgAhead.
|
| -s
SyncReleaseInodeLock
| Enables the code which minimizes the time spent holding inode lock during
sync by flushing dirty pages before calling _commit.
SyncReleaseInodeLock is a boolean variable: 0 to disable or a
positive integer to enable. The default is 0.
|
| -S
| Enables the SHM_PIN flag on shmget( ) system
call. By default this flag is ignored.
|
| -t maxclient
| Specifies the point above which the page stealing algorithm steals only
client file pages. This value is expressed as a percentage of the total
real-memory page frames in the system.
|
| -u lvm_Bufcnt
| Specifies the number of LVM buffers for raw physical I/Os. The
default value is 9. The possible values can range between 1 and
64. This option is only available in AIX Version 4.
|
| -U unixfile
| Specifies the name of the AIX file to patch for the -m and
-C flags. The default is
/usr/lib/boot/unix_mp.
|
| -w NpsWarn
| Specifies the number of free paging-space pages at which the operating
system begins sending the SIGDANGER signal to processes.
|
| -W MaxRandWrt
| Specifies a threshold (in 4KB pages) for random writes to accumulate in
RAM before these pages are sync'd to disk via a write-behind
algorithm. This threshold is on a per file basis.
The -W
MaxRandWrt option is only available in AIX 4.1.3 and
later. The default value of MaxRandWrt is 0, which disables
random write-behind.
|