lkml.org 
[lkml]   [1998]   [Jun]   [26]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: Thread implementations...
From"Michael O'Reilly" <>
Date26 Jun 1998 14:11:26 +0800
Rik van Riel <H.H.vanRiel@phys.uu.nl> writes:
> On Thu, 25 Jun 1998, Amsden, Zachary wrote:
> 
> > 	Possibly. But according to Squid's own doc, the primary problems
> > 	in Squid performance are
> > 
> > 	1) Not enough memory
> 
> Can indeed be alleviated by sendfile()...

No. The memory issue on squid is almost purely to do with needing to
keep a very large dbase (the index for all the disks) in ram.

> > 	2) Too slow disks
>

> With a filesystem that can't handle the extreme dir sizes
> that Squid uses.

Squid doesn't produce extreme dir sizes. No dir will have more then X
files for a given configuration.

> ReiserFS or another advanced filesystem
> will speed up this bottleneck by more than just a considerable
> amount...

The disk issue in squid is purely driven by transaction times.

> > 	which as far as I can see aren't helped by sendfile. They also
> > 	say "CPU limitations are rarely encountered except in very large
> > 	caches".
> 
> This situation is also better handled by tree-based filesystems
> and better buffer/cache administration.

On squid 1.2, large squids are CPU bound, and the number two CPU
sucker is read()s and write()s to network sockets. (squid does a LOT
of copying data from network to network, and disk to network).

Michael.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

\
 
 \ /
  Last update: 2005-03-22 12:43    [from the cache]
©2003-2008