On 4/24/07, Spencer Shepler <Spencer.Shepler at sun.com> wrote: > > On Apr 23, 2007, at 7:17 PM, Leon Koll wrote: > > > Hello, Roch > > <...> > >> Then SFS over ZFS is being investigated by others > >> within > >> Sun. I believe we have stuff in the pipe to make ZFS > >> match > >> or exceed UFS on small server level loads. So I > >> think your > >> complaint is being heard. > > > > You're the first one who said this and I am glad I'm being heard. > > > >> > >> I personally find it always incredibly hard to do > >> performance > >> engineering around SFS. > >> So my perspective is that improving the SFS numbers > >> will more likely come from finding ZFS/NFS > >> performance > >> deficiencies on simpler benchmarks. > >> > > There is a new version of SPEC SFS in beta phase (w/NFS4 and CIFS > > support), available to SPEC members only. I am very interested to > > see the results of it on ZFS. Is there anybody from [i]"others > > within Sun"[/i] who tested it ? > > The upcoming version of SPEC SFS, while containing support and > workload for CIFS, does NOT include support for NFSv4.
Sorry, my misunderstanding. > > I am also curious, Leon, does your regular workload > mimic that which is measured by the SFS workload or > are you just using it as a tool to determine variances > in your configurations? Unfortunately I don't know how well the SFS represents the regular workload. It's difficult to collect metrics of the current production NFS server (there is no DTrace because it's running Linux) so I am using SFS as a de-facto standard benchmark. > > Spencer > >