User: Password:
Subscribe / Log in / New account

KSM tries again

KSM tries again

Posted Apr 30, 2009 9:08 UTC (Thu) by simlo (guest, #10866)
Parent article: KSM tries again

Doesn't this in some ways replace/complement dynamic linking? It seems to be based on the same idea, but moves the whole job into the kernel.

Ofcourse, if two applications statically links to a library, there could be many reasons why the resulting pages within the executables are not identical :-(

(Log in to post comments)

KSM tries again

Posted Apr 30, 2009 9:45 UTC (Thu) by Darkmere (subscriber, #53695) [Link]

Not really, You think of this from a single-Machine perspective.

Think of it as this:
host machine running KSM:
VM 1:
VM 2:
VM 3:

In all these machines there is bound to be a lot of similar pages of memory allocated. For most (heh?) uses where this would be an improvement you run the same distribution+patchlevel on all VM's with some minor differences. In these cases things like prelink and memory randomization are in effect making the binaries different on disk, however, once you load them into RAM, the working state of for example /sbin/init is bound to have a LOT of pages in common between the four systems (host+3*VM).

And that is where you can get some neat memory savings.

Copyright © 2018, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds