Re: Slow NFS performance over wireless!

[Miquel van Smoorenburg]

In article <1074026758.4524.65.camel@xxxxxxxxxxxxxxxxxxxx>,
Trond Myklebust  <trond.myklebust@xxxxxxxxxx> wrote:
>There are a couple of performance related patches that should be applied
>to stock 2.6.0/2.6.1. One handles a problem with remove_suid()
>generating a whole load of SETATTR calls if you are writing to a file
>that has the "x" bit set. The other handles an efficiency issue related
>to random write + read combinations.
>
>Either look for them on my website (under
>http://www.fys.uio.no/~trondmy/src), or apply Andrew's 2.6.1-mm2 patch.

If one runs bonnie on a NFS mounted share, what should the rewrite
throughput be?

On an NFS server locally (2.6.1-mm3) I get as write/rewrite/read
speeds 107 / 25 / 110 MB/sec, CPU loads of a few percent.

On an NFS client (2.6.1-mm3, filesystem mounted with options
udp,nfsvers=3,rsize=32768,wsize=32768) I get for the same share as
write/rewrite/read speeds 36 / 4 / 38 MB/sec. CPU load is also
very high on the client for the rewrite case (80%).

That's with back-to-back GigE, full duplex, MTU 9000, P IV 3.0 Ghz.
(I tried MTU 5000 and 1500 as well, doesn't really matter).

Is that what would be expected ?

Mike.

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