Re: [squid-users] Squid approach to C10K problem

From: Kinkie <gkinkie_at_gmail.com>
Date: Tue, 17 Mar 2009 01:02:25 +0100

On Mon, Mar 16, 2009 at 6:08 PM, Roy M. <setesting001_at_gmail.com> wrote:
> Hi,
>
> Just out of curious, how does squid as a single threaded server, to
> handle massive amount of clients nowsday?
>
> http://www.kegel.com/c10k.html
>
> Especially some others (http://varnish.projects.linpro.no/) say squid
> is old fashion and don't honor those approach such as epoll/kqueue
> which is famous in handling many clients using minmount efforts.

Squid uses epoll and kqueue on Linux and FreeBSD, and has been doing
it for some time.
Squid 2.7 also supports Solaris' /dev/poll (squid 3.X doesn't) nor
MSWindows' Overlapped I/O and Completion Ports.
I suspect that the Varnish documentation is a bit out of date.

> But I know it is not true, I have been using Squid for years and think
> squid is very efficient and fast enought to serve massive clients. So
> I want to know if any hidden secrete behind?

It's hard to have any secrets when the code is in the open :P
Squid is reasonably efficient. It can be - and hopefully is being -
still improved.

> What is you guy comments on those issues ( c10k/epool etc)

Some of the most obvious issues have been addressed. There's still
lots to do tho.

-- 
    /kinkie
Received on Tue Mar 17 2009 - 00:02:31 MDT

This archive was generated by hypermail 2.2.0 : Tue Mar 17 2009 - 12:00:03 MDT