Re: General Runtime Information - Requests given to unlinkd

On Sat, Sep 19, 1998 at 09:37:26PM -0600, Alex Rousskov wrote:

> That is why I would like to see a convincing proof for the first
> two statements via a benchmark that simulates *Squid IPC traffic
> patterns* first.

I don't think you'll find one - because presently squid's IPC
patterns aren't that much of a problem.

Later on though, for much greater scalability I think they will be. I
mention this now because I think it is something that needs to be
considered in large MP boxes where the CPU isn't necessarily fast
enough to saturation either the memory bandwidth of the network.

-cw
Received on Tue Jul 29 2003 - 13:15:54 MDT