lkml.org 
[lkml]   [1999]   [May]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectLinux & NFS caching: reducing TCO
Date
Hi There,

in order to minimize the amount of time used for system maintenance, upgrading
etc. for our institute network (one main "file server" for /home, rest
"clients", near-complete 100 mbit installation) we thought about letting the
clients run diskless. A first test showed that this is quite easy to do and
really gives the benefit of one central installation for all clients.

However the performance, e.g. when using LaTeX, is quite awful. The next
thing we will try is upgrading the server from 2.0.x to 2.2.x in order to use
knfsd -- which should be somewhat faster. However the main issue seems to be
client-side caching. The normal mount parameters for nfs only seem to include
a timeout for attribute caching; using tcpdump to check how long a client
chaches file contents it seemed that that time never exceeds a couple of
seconds.

Is there any way to let the clients use a "proper" caching (even if just
for ro mounted /usr or so)? This could just make all the difference...

In such a case it might be necessary to let the server be able to send a kind
of invalidate message to the clients if something changes (I thought of keeping
data in cache for tens of minutes). This whole setup obviously relies on only
few changes and only makes sense for e.g. /usr but _not_ /home.

Regards, Colin Hirsch

--
Dipl.-Inform. Colin Hirsch hirsch@informatik.rwth-aachen.de
Research Group on Mathematical Foundations of Computer Science
University of Technology Aachen Germany
--

-
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/

\
 
 \ /
  Last update: 2005-03-22 13:51    [W:0.721 / U:0.044 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site