there's also a technical report that describes work we did recently to
improve write performance, which summarizes the web page Trond listed
below. see
http://www.citi.umich.edu/techreports/reports/citi-tr-01-12.pdf
we're publishing a revised version in the June 2002 Usenix technical
conference proceedings; if you intend to cite the paper, we'd prefer that
you cited that one instead.
On Tue, 12 Feb 2002, Trond Myklebust wrote:
> >>>>> " " == Dave Hansen <haveblue@us.ibm.com> writes:
>
> > Trond Myklebust wrote:
> >> The following patch strongly reduces BKL contention within the
> >> NFS read/write code, and within the generic RPC layer.
>
> > Do you have any benchmarks which showed BKL contention in the
> > NFS code?
> > I'm not trying to criticize, I think the patch is wonderful.
> > I want
> > to have some more numbers to say, "Look! The BKL _is_ bad!"
>
> See Chuck's paper on
>
> http://www.citi.umich.edu/projects/nfs-perf/results/cel/write-throughput.html
>
>
> Cheers,
> Trond
>
- Chuck Lever
-- corporate: <cel@netapp.com> personal: <chucklever@bigfoot.com>- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Feb 15 2002 - 21:00:50 EST