Re: Performance question

On Tue, 15 Jun 1999, Simon Rainey <srainey@rmplc.net> wrote:
>As a test of Squid's potential I configured it to not cache anything to
>disk and set it up to use the Peregrine proxy as a parent. Lo and behold,
>Squid could more than saturate Peregrine, and this is using real request
>patterns in a live network.
>
>I think everyone realises that Squid's achilles heal is the fact that it
>relies on the undelying OS to manage disk files. While the tests I've
>done are admittedly not very scientific, they do suggest that effort
>spent on the rumoured Squid-FS (or similar) would be worthwhile.

Hmm. You mentioned that you're running with 18 servers (presumably
siblings)? If the FS is truly the limiting factor, it may be that running
them all with proxy-only would help things out. The basic notion being
that it would probably be faster to get something out of a sibling's
in-memory cache than to get it off your disk.

The downside is that you might end up getting it off of a sibling's disk.
But, hopefully, the expanded in-memory cache may offset that loss.

Later,
scott
Received on Wed Jun 16 1999 - 10:05:03 MDT