Re: [squid-users] Frequent cache rebuilding

From: Adrian Chadd <adrian_at_squid-cache.org>
Date: Thu, 22 Jan 2009 14:04:05 -0500

2009/1/21 Amos Jeffries <squid3_at_treenet.co.nz>:

> Yes it can. Squid's passing through of large objects is much more
> efficient than its pass-thru of small objects. A few dozen clients
> simultaneously grabbing movies or streaming through a CONNECT request can
> saturate a multi-GB link network buffer easily.
> A dual-core Xeon _should_ be able to saturate a 10GB link with all clients
> online.

Has anyone tried this?

The last time I tried multi-gige with Squid, it didn't really hit
anywhere near 10GE with streaming data because current kernels are
optimised with the idea that people will write concurrent software,
and so will run multiple threads to do the socket and network stuff
(copyin/copyout/tcp/ip stuff, with a kernel thread handling part of
the NIC stuff and potentially some of the TX/RX.)

Adrian
Received on Thu Jan 22 2009 - 19:04:14 MST

This archive was generated by hypermail 2.2.0 : Fri Jan 23 2009 - 12:00:02 MST