2009/8/2 Heinz Diehl <htd_at_fancy-poultry.org>:
> 1. Change cache_dir in squid from ufs to aufs.
That is almost always a good idea for any decent performance under any
sort of concurrent load. I'd like proof otherwise - if one finds it,
it indicates something which should be fixed.
> 2. Format /dev/sdb1 with "mkfs.xfs -f -l lazy-count=1,version=2 -i attr=2 -d agcount=4"
> 3. Mount it afterwards using "rw,noatime,logbsize=256k,logbufs=2,nobarrier" in fstab.
> 4. Use cfq as the standard scheduler with the linux kernel
Just out of curiousity, why these settings? Do you have any research
which shows this?
> (Btw: on my systems, squid-2.7 is noticeably _a lot_ slower than squid-3,
> if the object is not in cache...)
This is an interesting statement. I can't think of any specific reason
why there should be any particular reason squid-2.7 performs worse
than Squid-3 in this instance. This is the kind of "works by magic"
stuff which deserves investigation so the issue(s) can be fully
understood. Otherwise you may find that a regression creeps up in
later Squid-3 versions because all of the issues weren't fully
understood and documented, and some coder makes a change which they
think won't have as much of an effect as it does. It has certainly
happened before in squid. :)
So, "more information please."
Adrian
Received on Sun Aug 02 2009 - 11:01:51 MDT
This archive was generated by hypermail 2.2.0 : Tue Aug 04 2009 - 12:00:03 MDT