> First results look good.
> Using 2.2.5-ac4 with the patches you distributed everything
> compiled clean.
>
> FINALLY, I can write from a linux machine via nfs to a Solaris server
> at right around 2 megabytes per second (100MB Lan). This is acceptable
> since before vers 3 support writing was around 500k/s at best.
>
> However, reading has dropped to around 500k/s. Can't we win?
How does the current version perform (0.5.2)? On my setup, reading is
back up to scratch (yes, it did dip a while) and write performance is
slightly better.
> I mounted the filesystem with options nfsvers=3,rsize=8192,wsize=8192.
>
> Don't know how safe it is, but I tried wsize of 32768 and was writing
> files between 3 megabytes and 8 megabytes per second! Changing the
> rsize to this didn't make too much difference. Changing the rsize to
> 1024 did slow it down by a factor of 4. Is it possible the rsize is
> stuck at 4096 or something?
Yes. We currently cluster writes in blocks of wsize and then shove
them out in one fell write, however we do not do this for reading. The
reason is that we already have the file readahead feature, so the gain
would not be as great. The clustering would, however, be easy to adapt
to provide read clustering, so if it turns out useful, I could extend
it.
Cheers,
Trond
NB: The current 'ac' release is against 2.2.5-ac6. I've just seen ac7
on ftp.no.kernel.org, but I can't be bothered to update that just this
minute. I assume Alan hasn't changed much to the NFS stuff, so it will
probably patch cleanly.
Oh: and don't forget to update 'mount' as well. My previous release
had some bugs. My humble thanks to those who helped me out with patches...
-
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/