Re: how can I study effectiveness of cache BEFORE installing?

From: Andrew Brennan <>
Date: Wed, 4 Mar 1998 11:47:41 -0500 (EST)

   You can't lose by installing it and using it (assuming you have the
   hardware & time resources available), so set it up and configure it
   for a sample group -- preferrably of tech folk. If you can see the
   benefit in the log files (HITs vs. MISSes, etc.) and the server has
   improved access times ... those are the numbers you want, no?

   This is what I've done locally (although the end users have seen a
   visible performance boost and don't need any numbers) to shift some
   of our user population from a (poorly-run) Microsoft Proxy server.

   A sniffer can measure HTTP activity to get a sense of the numbers,
   but it's not going to be able to give you accurate numbers like the
   logs of a well-configured Squid system. You *could* use the output
   of the sniffer to measure the difference in traffic levels before &
   after ... but this would only be for your sample group and requires
   that you actually set up a Squid system.

On Wed, 4 Mar 1998, Bob Bagwill wrote:

> I would like some way of know how much benefit we would get
> from a cache without installing it and making people use it.
> Has anyone used a sniffer or equivalent to pass HTTP to
> a squid server to study the cache effectiveness? Thanks.
Received on Wed Mar 04 1998 - 08:49:43 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:39:09 MST