Re: squid2.3stable2 on freebsd3.2

Chris Dillon wrote:

> Ok, this makes sense. :-) It might, in fact, make even more sense for
> FreeBSD 4.x because of the improved VM system.

Well, actually not.

The intention of memory pools is to limit fragmentation and thereby
reduce Squids VM usage. The problem is that the default configuration
puts no limit on the amount of idle memory in the pools..

> I'm not sure that this rather small performance improvement will be
> realized when Softupdates is involved, but that would involve another
> benchmark to find out. :-)

Theoretically untruncate causes less disk I/O due to directories and
inode bitmaps not being updated once the cache has filled (twice).

--
Henrik Nordstrom
Squid hacker