[NFS] New version of Linux NFSv3 client is out...

[Michael Clark]

Todd Chauvin wrote:
> 
> >
> > Hello,
> >
> >   I've put up a new version of the Linux NFSv3 client in the
> > usual place. Please find it as
> >
> >   http://www.fys.uio.no/~trondmy/src/linux-2.2.14-nfsv3-0.16.11.dif.bz2
> >
> 
> Hi.
> 
> I thought that the readership of the list might be interested in some
> quick results with Trond's latest v3 client against a solaris7 server.
> Note that these are from a single run over a network that was quiet but
> not completely without other traffic.
> 
> All tests:  solaris7 server, numbers produced by bonnie with filesize 50 mbytes.
> The network is 10mbits/sec ethernet.  The server and client are separated by
> both a passive hub and a smart switching hub.
> 
>               -------Sequential Output-------- ---Sequential Input-- --Random--
>               -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
> Client        K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %CPU
> 2.2.14(v2)     245  3.4   156  0.2   352  0.6  7535 99.9 254938 99.6 173.3  0.9
> 2.2.14(v3-udp) 690 10.1   705  1.0   567  1.2  7337 99.3 250028 97.7 613.1  2.9
> 2.2.14(v3-tcp) 582  8.5   861  0.7   865  1.0  7373 99.8 254065 99.2  46.7  0.4
> solaris7       922 16.3   851  2.1   523  1.8   919 13.1    947  1.6  26.1  2.1

                                                           ^^^^^^
What's the story with the 250Mb/sec sequential block rate? Is the linux
client caching this data? Should it really be doing this?

Other figures looks great.

> Even with these simple numbers (with admittedly high variance), I can
> see a dramatic improvement as compared to the stock 2.2.14 v2 client.
> For reference, I've included as the last line the performance of a
> solaris7 client to the same server through the same network hubs as the
> linux machines.
> 
> When I first started using linux nfs, the thoughput to solaris was a mere
> 50 K/sec!  Thanks to everyone for the hard work that has made these
> dramatic improvements possible!
> 
> Todd Chauvin
> chauvin en ee.washington.edu
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo en vger.rutgers.edu
> Please read the FAQ at http://www.tux.org/lkml/

-- 
  ~mc <michael en astrolog.net>

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