Re: some linux tuning (fs benchmarks)

> system benchmarking issue. The Bonnie benchmark does operations on a
> *single* file. If that file fits entirely in your system's RAM, all

Doh, I'm sorry! I *completely* had some hosed thinking all along!

> will dominate. That's why Alex Rousskov said the benchmark needs to
> fit Squid's usage patterns (though I still think it doesn't need to be
> specific to Squid - lots of apps have at least roughly that kind of
> macro usage pattern.)

Sure, I was thinking along the lines of usage patterns in an earlier
post. With Squid having any given amount of file descriptors actually
active, Bonnie could never truely hack this.

Reiterating my earlier goal, I just want to be able to say "Processor X
with HBA Y running RAID 0 is a great performer!" If 'md' can surmount a
RAID controller, I'd like to know.

Are we on to Polygraph, or do you know of some existing case studies that
might be able to give us what we(?) want?

I can recover from this... :)
Received on Mon Jun 28 1999 - 17:00:18 MDT