Re: some linux tuning (fs benchmarks)

Alex Rousskov wrote:

> On Mon, 28 Jun 1999, Clifton Royston wrote:
>
> > I think that Doug's suggestion was to measure file system performance
> > independently, to be able to understand it as a limiting factor in
> > Squid or other proxy performance.
>
> Sure. My point was that unless the proxy-independent test mimics Squid
> file system usage patters, it is pretty much useless (you are not
> getting an asymptotic behavior, you are getting some constant that is
> most likely not relevant to Squid). To produce useful results, one has
> to go through Squid with a micro-benchmark or write a proxy-independent
> benchmark from scratch.
>
> Alex.

Agreed. There is one person I have to deal with who cannot seem to grasp that
disk seek latency and controller bandwidth doesn't tell the whole story when
it comes to cache-read/write speeds and overall proxy performance. I guess we
should be flattered, therefore, that he treats the operation of all the
software as 'perfect'.

D
Received on Mon Jun 28 1999 - 17:38:12 MDT