NFS Read Performance

we're conducting SPEC SFS97 benchmark runs to evaluate Linux performance as NFS file servers. It is observed that an unusual high percentage of benchmark time was spent in "read" operation. A sampled workload consisting of 18% of read consumes 63% of total benchmark time. Did this problem get analyzed before (or even better :)- is there a patch) ? We're on 2.4.19 kernel- NFS V3 - UDP, with EXT3 as local file system.

Thanks in advance.

-------------------------------------------------------
This SF.net email is sponsored by:Crypto Challenge is now open! Get cracking and register here for some mind boggling fun and the chance of winning an Apple iPod:
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs