How to measure performance?

Peter J. Holzer hjp at
Mon Oct 4 15:19:18 GMT 1999

On 1999-10-04 08:39:05 -0400, David Collier-Brown wrote:
> Peter J. Holzer  wrote:
> | This is not enough. Many applications do not read and write files
> | sequentially. 
> 	Yes, but that's probably too much to ask Ms. Vologdin
> 	to do on the first go-round... 

Ok, maybe not for a first go-round, but one should be prepared to see
significantly lower performance in real life than in simplistic
benchmarks. The recipy given sounds like it could build good traces for
a benchmark, but I seriously doubt it does.

> | his is evident for MS-Access (which reads and writes
> | single records), 
> 	I'm on the record as recommending one never try to run
> 	a multi-user database against a remote filesystem: search
> 	for "free sample of crack cocaine" in the archives (;-))

I agree completely, but our users don't :-(

> | There are test suites like the BapCo suite which run the actual
> | applications (but we couldn't find much difference between a local
> | disc and one shared over 10 Mbit ethernet, so at least Bapco is
> | probably more CPU-bound).
> 	Actually, oplocks cause local caching of the files,
> 	so the non-sequential accesses are mostly to local disk.
> 	That's possibly what fools bapco.

Actually, the tests we did were against 1.9.16, which didn't support
oplocks (Don't ask why we haven't upgraded to a newer version yet - it's

> 	[Andrew has a newish load-generation suite, specifically for
> 	SMB: I expect folks will eventually start using and reporting
> 	based on it...]

Sounds interesting. Is it on the samba web site? I couldn't find it.


   _  | Peter J. Holzer             | Nobody should ever have to be
|_|_) | Sysadmin WSR / Obmann LUGA  | ashamed if they have a secret love
| |   | hjp at               | for writing computer programs that
__/   | | actually work.  -- Donald E. Knuth
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 371 bytes
Desc: not available
Url :

More information about the samba mailing list