[OpenAFS] Extremely poor write performance.

Rubino Geiß kb44@rz.uni-karlsruhe.de
Fri, 17 Jan 2003 17:15:33 +0100


> On Fri, Jan 17, 2003 at 09:58:46AM +0100, Rubino Gei? wrote:
> > 	1. guess: maybe it's the network latency. 
> > 		- contraidication: on 10Mbit we got the same 
> numbers but 5 ro 10 
> > times higher ping turnarounds.
> > 	2. guess: a flaw in the server code or linux.
> > 		- guestion: how can I profile the fileserver? I 
> want to know where 
> > he is consuming time / waiting. How is this done for a client?
> 
> what about too many interrupts ?
> How is the underlying filesystem performance ? might the 
> problem be there ? Test it on the server and see what you get.
> JonB

on the ext3-fs-partition (for the afs-file-server):

        ---Random Create---
        -Create-- -Delete--
 files  /sec %CP  /sec %CP
    16  1434  97  3333  82

on afs with memcache:

        ---Random Create---
        -Create-- -Delete--
 files  /sec %CP  /sec %CP
    16   233  25   209   8

on afs with diskcache:

        ---Random Create---
        -Create-- -Delete--
 files  /sec %CP  /sec %CP
    16   217  28   198  10

>>>>>> Conclusion: ext3 ist fast enough, and memcache / diskcache cannot be
the rescue.