[GRLUG] large systems performance

[Michael Mol]

On Nov 16, 2007 1:21 AM, Tim Schmidt <timschmidt at gmail.com> wrote:
> On Nov 16, 2007 1:16 AM, Michael Mol <mikemol at gmail.com> wrote:
> > Heh.  It could start hurting even more.  Apparently they're getting
> > close to implementing swap over NFS.  It's intended for diskless
> > cluster environments, but it still stands to complicate performance
> > issues.  Sub-optimal performance?  Do you have a bad network cable, or
> > are your jobs eating too much memory?
>
> Swap over NFS has worked for years...  LTSP has (optionally) supported
> it for ages.

Not perfectly; The NFS layer may require memory to be allocated, which
can be a problem if you're calling the NFS layer to get access to swap
so that you can free up some memory to honor an allocation request.
If there's already a lack of memory, the allocation request can fail.

Linky:
http://kerneltrap.org/Linux/Swap_Over_NFS


-- 
:wq