Re: Bad NFS performance with client 2.2.9

Trond Myklebust (trond.myklebust@fys.uio.no)
Mon, 31 May 1999 14:17:38 +0200 (CEST)


>>>>> " " == Steven N Hirsch <shirsch@adelphia.net> writes:

> On Wed, 26 May 1999, Trond Myklebust wrote:
>>
>> Writing the 120 Megabyte file, 'zz'...13.900000 seconds Reading
>> the file...14.120000 seconds
>>
>> IOZONE performance measurements: 9052454 bytes/second for
>> writing the file 8911410 bytes/second for reading the file

> That's really smoking! Is this using the NFS v3 protocol? I
> am lucky to see half that between my boxes (over 100Base
> ethernet).

> I had tried NFS v3 at one point, but started seeing some scary
> messages in the logs and chickened out <g>.


That's actually using the NFSv2 protocol with the latest NFSv3 patches
(I've done some work on polishing up the sunrpc stuff). There were a
lot of cases in which the errors weren't reset, the timeouts weren't
being set properly before sleeping and the like. If it cures Thierry's
problem, and proves stable enough I'll pass the patch on to Alan.

Cheers,
Trond

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/