<br><font size=2 face="sans-serif">That's basically the answer that I got
from an original developer: "we were actively considering support
for NFS and GPFS, both have client side caching".</font>
<br>
<br>
<br><tt><font size=2>Rob Ross <rross@mcs.anl.gov> wrote on 02/08/2008
08:39:34 PM:<br>
<br>
> My guess is that it's the "lowest common denominator", that
it works <br>
> for both GPFS (albeit with unnecessary locking calls) and for NFS,
<br>
> which are the two configurations for which IBM is interested.<br>
> <br>
> Rob<br>
> <br>
> On Feb 8, 2008, at 5:40 PM, Bob Cernohous wrote:<br>
> <br>
> ><br>
> > robl@mcs.anl.gov (Robert Latham) wrote on 02/08/2008 04:22:17
PM:<br>
> ><br>
> > ><br>
> > > Do you know why ad_bgl is based off of ad_nfs and not ad_ufs?
The<br>
> > > aggressive locking in ad_nfs is an attempt to flush client-side<br>
> > > caches, but shouldn't be needed for other file systems.<br>
> ><br>
> > No, sorry. I'm pretty new to romio. I'll try to find
some answers<br>
> > (and educate myself on <br>
> > this)._______________________________________________<br>
> > dcmf mailing list<br>
> > dcmf@lists.anl-external.org<br>
> > http://lists.anl-external.org/cgi-bin/mailman/listinfo/dcmf<br>
> > http://dcmf.anl-external.org/wiki<br>
> <br>
</font></tt>