Re: some linux tuning (fs benchmarks)

On Mon, 28 Jun 1999, Doug Renner wrote:

> FS speed, specifically I/O bandwidth, is uberimportant to caches.

I/O bandwidth is not a bottleneck for most Squid setups. File system
overheads, blocking processes, no application level seek optimization,
etc. are.
 
> Unless we already have a standard benchmark, I propose that we create
> one. I would like to see published results in some standard, moderately
> trustable, format. Maybe there is already a project that has FS specific
> benchmarks. I would like to see them with these different hardware
> combinations.

Testing raw disk performance or raw file system performance seems
counter-productive to me. You can use Web Polygraph to test disk
subsystem of a proxy with Web-like workload:
        http://polygraph.ircache.net/

We already have published benchmark results for several vendors and more
results are underway:
        http://polygraph.ircache.net/Events/

However, the results that we publish are not micro-level benchmarks
because some vendors are afraid of those. Note that one cannot publish a
benchmark figures for a commercial proxy without explicit permission.

You can certainly come up with a Poly workload to benchmark disk
subsystem and then encourage people to run it against their Squids. We do
that kind of tests when testing new Squid features:
        http://squid.nlanr.net/Squid/Benchmarking/

Alex.
Received on Mon Jun 28 1999 - 12:25:59 MDT