lkml.org 
[lkml]   [2000]   [Mar]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Overcomittable memory (Was: Linux 2.2.15pre12)
On Mon, 13 Mar 2000, Jamie Lokier wrote:

> Khimenko Victor wrote:
> > And my 20 apache processes will require 800MiB of virtual memory instead of
> > 60MiB they are using now.
>
> If those Apaches _might_ conceivably all enter mod_perl at the same time
> and run a tricky script, and you want to _guarantee_ that nothing can
> die, 800MiB is the correct vm requirement.

No. That's the requirement if they all "dirty" every single page of their
own image. Shall we try including the RAM required to lock all the
mmap()ed pages into physical RAM at once, too? How about requiring
physical RAM to be bigger than physical RAM + swap space + filesystems?

Besides, this is all wrong anyway - see below.

> Your choice. Use 60MiB and accept that there are conditions under which
> the Apaches can die, or buy a bigger disk and claim that they cannot.

No. Those theoretical conditions you speak of would require Apache and
mod_perl to be self-modifying code on an appalling scale.

> (Heavily swapping Apaches are probably worse than dead ones, but that's
> your choice to make).

Since neither is really a possibility, though, this is just silly.

If you REALLY want to allow for the worst case: Each process has an
address space of 3 Gb (+1Gb kernel). Your method would require, for a
sytem with a hard limit of 1024 processes, a minimum of 3Tb swap+RAM. You
can try this if you want, but I don't think many people will follow this.


James.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:57    [from the cache]
©2003-2014 Jasper Spaans. Advertise on this site