[squid-users] Re: Performance problems - need some advice

From: Jan-Frode Myklebust <mykleb@dont-contact.us>
Date: Wed, 8 Feb 2006 10:42:15 +0100

On 2006-02-07, Gregori Parker <gregori@playstream.com> wrote:
>
> Can anyone recommend techniques for assessing squid performance? I
> have no good way of benchmarking our clusters since SNMP isnt ready
> quite yet.

I use Performance CoPilot from http://oss.sgi.com/projects/pcp/ for
keeping track of squid and server performance. It comes by default
with a huge number of system performance metrics, and also has a nice
plugin (PMDA, Performance Metrics Domain Agent) for collecting metrics
from the squid access.log.

i.e. it can collect historic, or show live how many requests/s or
byte/s squid is answering of type:

        total
        get
        head
        post
        other
        size.zero le3k le10k le30k le100k le300k le1m le3m gt3m unknown
        client.total
        cached.total
        cached.size.zero le3k le10k le30k le100k le300k le1m le3m gt3m unknown
        uncached.total
        uncached.size.zero le3k le10k le30k le100k le300k le1m le3m gt3m unknown

and also combine this with system level metrics like load, system cpu
time, cpu i/o wait, per partition byte/s, network interface byte/s,
and much more..

Because of it's historic logs of all this, it's great for collecting
the performance numbers during high activity, and then replaying it
to analyse what goes wrong later on.

  -jf
Received on Wed Feb 08 2006 - 02:43:23 MST

This archive was generated by hypermail pre-2.1.9 : Wed Mar 01 2006 - 12:00:03 MST