Benchmark results

Jeremy Allison jallison at
Wed Feb 3 20:09:24 GMT 1999

David Collier-Brown wrote:
>   Methinks this benchmark allows lots of server-side caching (;-))
> Anyone care to comment on the method(ology)?

No - it's actually client side caching (the oplocks, see).

What NetBench actually measures in a large part is the
speed of the client memory cache :-) :-) :-).

This is why we beat up ZDNet at the last CIFS conference.

NetApp's proposed CIFSbench isn't any better as it's a
proprietary Win32 based app.

The only reasonable benchmark would be one based on
open code that generated the CIFS calls via direct
socket calls - not going through a Windows redirector
(which is broken in *so many* subtle and not-so-subtle
ways :-). The smbtorture code would be a good start

It still doesn't stop me trumpeting the Samba Netbench
figures when I get good ones though - that's called
"marketing" :-) :-) :-).



Buying an operating system without source is like buying
a self-assembly Space Shuttle with no instructions.

More information about the samba-technical mailing list