[lkml]   [1999]   [May]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: Linux & NFS caching: reducing TCO
On 13 May 1999, Trond Myklebust wrote:

Colin Hirsch <> writes:

> 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.

That would not be NFS. NFS is stateless: the server neither knows nor
cares what the clients are caching.

Feel free to invent your own protocol, but beware that the sort of
thing you're proposing will most likely place a very heavy burden on
both the server and your network.


Slightly off-point, but..

I remember there was an nfs hack whereby you could have different
files going to different clients from the same nfs volume based on an
extension of the filename. eg something like:

ls on the server of nfs export:

size 2242 file1:-client1
size 3091 file1:-client2

ls from client1:

size 2242 file1

ls from client2:

size3091 file1

ie, the server exports different file1's to different clients based
on the extension.

I remember reading about this somewhere in the linux/Documentation
directory a long time ago. But I can't find back anymore now, just
when it'd be useful to me.

Anybody remember this feature? and whether it still exists?

Paul Jakma
PGP5 key:
There's got to be more to life than compile-and-go.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:51    [W:0.183 / U:1.488 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site