[squid-users] Vedr.: Re: [squid-users] squid performance of 2.4 much worse than 2.2 and 2.3?

I hope squid-2.5 is as fast as 2.2-stable5.. - I like the rproxy features.
a lot.

Do you guys have any performance pointers, as to how and what I need to
tune.. both on the squid.. and on the linux-box?

I read something about changing tcp_rmem and tcp_wmem so they can open up
bigger tcp-windows.. I did that..
but the cache mainly serves files in size of 10 - 110K.. and as the normal
window on a linux-2.4 has a max of 256kb that shouldn't be a problem?

how many aio processes do I need to run on a busy box such as this? - i
started with 50.. I tried with 350..
can't seem to see if that helps at all..

As you can see from my vmstat output.. it was the interrupts that seemed to
be the performance stopper.. it maxed out at approx. 18.000 pr. second -
according to vmstat output (see other email).

I guess the interrupts are for the network? - unfortunately I can't see
from netstat -s - how many of the resends there are - are from that box..

But I do know this.. - while squid was freezing - and the vmstat output
that you saw ran.. I could easily connect via ssh - no problems or delays..
(that's how I got the vmstat output :-)

That seems to me, that it wasn't a network problem.. and if the interrupts
that seemed to - hit the ceiling at 18.000.. wasn't due to network
interrupts.. then what were they caused by? - the disks had almost no
load.. according to the same vmstat output..

Have any ideas guys? - to me it seems fishy...

-------------| This mail has been sent to you by: |------------
Klavs Klavsen, IT-coordinator and Systems Administrator at
Metropol Online - http://www.metropol.dk
Tlf. 33752700, Fax 33752720, Email ktk@metropol.dk

Private- Email klavs@klavsen.net - http://www.vsen.dk

--------------------[ I believe that... ]-----------------------
It is a myth that people resist change. People resist what other
people make them do, not what they themselves choose to do...
That's why companies that innovate successfully year after year
seek their peopl's ideas, let them initiate new projects and
encourage more experiments. -- Rosabeth Moss Kanter

Squid-2.2.STABLE5-hno is proven quite fast..

Squid-2.3 has some problems.. especially in it's async_ufs/asyncio
implementation.

Squid-2.4 fixes most

Squid-2.5 is told to be quite okay.. but there is no hard proofs yet.
There is a Backeoff starting today, and then we will see how it
compares...

Regards
Henrik Nordström
Squid Hacker

Klavs Klavsen wrote:
>
> Hi Gary and everyone else on the squid-users list..
>
> Is Gary right in his understanding that squid 2.2 and 2.3 are
considerably
> faster than the current 2.4 stable?
> and whatabout the 2.5?
>
> best regards,
> Klavs
>
> Hello Klavs,
>
> Have you tried one of the older versions of squid, like 2.2 or 2.3? i
> understand that the performance for 2.4 is considerably less than the
> older versions. of course, if you need features that are only available
> in one of the newer squid versions, i suppose this wouldn't help much.
Received on Tue Nov 13 2001 - 00:26:04 MST