[OpenAFS] Extremely poor write performance.

[Martin Schulz]

Paul Blackburn <mpb@est.ibm.com> writes:

> It seems that you looked at the performance of 1 client with 1 server?
> Are your figures for a single access to a server?

That is the natural measuring a user would do. That is what is
important to *him*. That is what *he* will judge AFS on. 

He will more often than not write to a single volume
and the writes seem to be a problem, as the subject suggests. 


You describe what an admin would judge AFS on:

> What would be more interesting than a single-server:single-client
> comparison  is to look at how performance varies as the number
> of clients increases.

Well, I am just wondering: What about concentrating "an increasing
number of clients" on a single machine?

Practically speaking: What happens if you drastically increase the
number of afsd daemons (to say 300 instead default of 3 for "Medium"
in the /etc/sysconfig/afs file)?

How does this affect the "delete 10000 small files" performance? 

Robin and Rubino, could you try this and report, please ? 

Yours,
-- 
Martin Schulz                             schulz@iwrmm.math.uni-karlsruhe.de
Uni Karlsruhe, Institut f. wissenschaftliches Rechnen u. math. Modellbildung
Engesser Str. 6, D-76128 Karlsruhe