KSM (kernel same-page merging) - Revision history

Corradofiore: Corrected typo (KSM should be written in lowercase)

2016-10-10T04:35:05Z

Corrected typo (KSM should be written in lowercase)

Corradofiore: Added note in case yum-downloadonly plugin is not found

2016-10-10T04:34:11Z

Added note in case yum-downloadonly plugin is not found

77.41.63.43: export LD_PRELOAD

2016-07-02T20:51:11Z

export LD_PRELOAD

Sergey Bronnikov: Marked this version for translation

2015-12-26T08:40:33Z

Marked this version for translation

Show changes

Sergey Bronnikov: add 'translate' tags

2015-12-25T22:49:31Z

add 'translate' tags

Corradofiore: /* Tuning */ Fixed a typo in the command: there was a ":50" at the end.

2015-07-21T13:35:13Z

‎Tuning: Fixed a typo in the command: there was a ":50" at the end.

Kir: added to HOWTO

2015-07-20T21:41:25Z

added to HOWTO

Kir: Kir moved page KSM (Kernel Samepage Merging) to KSM (kernel same-page merging)

2015-07-20T21:41:04Z

Kir moved page KSM (Kernel Samepage Merging) to KSM (kernel same-page merging)

Kir: added references and external links sections; minor rewording in intro

2015-07-20T21:40:21Z

added references and external links sections; minor rewording in intro

Corradofiore: /* Caveats */

2015-07-18T11:29:22Z

‎Caveats

@@ Line 113: / Line 113: @@
 == How to check efficiency of KSM == <!--T:26-->
 To check if KSM is actually reducing memory usage, issue this command on the HN:
-  [root@HN /]# cat /sys/kernel/mm/KSM/pages_sharing
+  [root@HN /]# cat /sys/kernel/mm/ksm/pages_sharing
 If the value is greater than 0, you're saving memory.  Refer to https://www.kernel.org/doc/Documentation/vm/ksm.txt for more details.

@@ Line 77: / Line 77: @@
   [root@container /]# yum install -y yum-downloadonly
       ''...bunch of output...''
-−
   [root@container src]# yum -y --downloadonly --downloaddir=/usr/local/src install ksm_preload
       ''...bunch of output...''

← Older revision		Revision as of 20:51, 2 July 2016
Line 101:		Line 101:

	'''LD_PRELOAD=/usr/lib64/libksm_preload.so'''		'''LD_PRELOAD=/usr/lib64/libksm_preload.so'''
		+	'''export LD_PRELOAD'''

	''mode=$1 # start or stop''		''mode=$1 # start or stop''

@@ Line 1: / Line 1: @@
 KSM is a memory-saving de-duplication feature, developed by Red Hat. It first appeared in the Linux kernel version 2.6.32.
@@ Line 111: / Line 112: @@
 * [[wikipedia:Kernel same-page merging]]
+</translate>
 [[Category: HOWTO]]

@@ Line 99: / Line 99: @@
 On a production machine, you'll want to modify some of the default values.  A more sane value for <code>/sys/kernel/mm/KSM/sleep_millisecs</code> is usually between 50 and 250 (YMMV though):
-  [root@HN ~]# echo 50 > /sys/kernel/mm/ksm/sleep_millisecs:50
+  [root@HN ~]# echo 50 > /sys/kernel/mm/ksm/sleep_millisecs
 == Caveats ==

← Older revision		Revision as of 21:41, 20 July 2015
Line 111:		Line 111:

	* [[wikipedia:Kernel same-page merging]]		* [[wikipedia:Kernel same-page merging]]
		+
		+
		+	[[Category: HOWTO]]

← Older revision	Revision as of 21:41, 20 July 2015
(No difference)

@@ Line 1: / Line 1: @@
-KSM is a memory-saving de-duplication feature, added by RedHat to the Linux kernel version 2.6.32.
+KSM is a memory-saving de-duplication feature, developed by Red Hat. It first appeared in the Linux kernel version 2.6.32.
 KSM replaces RAM pages of identical content with a single write-protected page, which in turn gets automatically copied to a new one if a process later wants to update its content.  This makes the de-duplication mechanism transparent to applications.  This strategy is commonly known as COW (''Copy On Write'').
@@ Line 102: / Line 102: @@
 == Caveats ==
-The ksmd daemon will take one or two minutes to start deduplicating memory and will require several minutes to reach stable state.  During the boot phase your HN could start swapping if you have heavily overcommitted your RAM.  You might want to use more aggressive settings (higher <code>pages_to_scan</code>, lower <code>sleep_millisecs</code>) at the beginning, effectively trading CPU utilisation for less chances of disk swapping, and then relax them after 10 mins or so.  Another possibility is to place your swap onto an SSD drive.
+The ksmd daemon will take one or two minutes to start deduplicating memory and will require several minutes to reach stable state.  During the boot phase your HN could start swapping if you have heavily overcommitted your RAM.  You might want to use more aggressive settings (higher <code>pages_to_scan</code>, lower <code>sleep_millisecs</code>) at the beginning, effectively trading CPU utilization for less chances of disk swapping, and then relax them after 10 mins or so.  Another possibility is to place your swap onto an SSD drive.

← Older revision		Revision as of 11:29, 18 July 2015
Line 102:		Line 102:

	== Caveats ==		== Caveats ==
−	The ksmd daemon will take one or two minutes to start deduplicating memory and will require several minutes to reach stable state. During the boot phase your HN could start swapping if you have heavily overcommitted your RAM. You might want to use more aggressive settings (higher <code>pages_to_scan</code>, lower <code>sleep_millisecs</code>) at the beginning and then relax them after 10 mins or so. ~~Others suggest~~ to place your swap onto an SSD drive.	+	The ksmd daemon will take one or two minutes to start deduplicating memory and will require several minutes to reach stable state. During the boot phase your HN could start swapping if you have heavily overcommitted your RAM. You might want to use more aggressive settings (higher <code>pages_to_scan</code>, lower <code>sleep_millisecs</code>) at the beginning, effectively trading CPU utilisation for less chances of disk swapping, and then relax them after 10 mins or so. Another possibility is to place your swap onto an SSD drive.