[Wien] parallel calc. without NFS

[Kevin Jorissen]

I started to use local SCRATCH directories on our machines some months ago
and may be able to help you if you run into problems there (some scripts
need to be rewritten).

Another thing you may do is add delays in lapw1para and/or lapw2para (as
Stefaan showed me some time ago) : focus on the sleep and delay parameters
in the lapw1para and lapw2para scripts, or add extra instructions.

Kevin.

----- Original Message -----
From: "Stefaan Cottenier" <Stefaan.Cottenier at fys.kuleuven.ac.be>
To: <wien at zeus.theochem.tuwien.ac.at>
Sent: Wednesday, July 09, 2003 7:49 AM
Subject: Re: [Wien] parallel calc. without NFS


>
> > Is there any way of running either k-point parallel or MPI versions of
> > Wien2k on a cluster of computers without using NFS? i.e., only using
local
> > hard disks of each node.
> >
> > Our NFS is agonisingly slow and wien2k reads & writes often hang and
data
> > gets corrupted. If anyone has a hack for getting around this I'd
> > appreciate it.
>
> What do you use as $SCRATCH ?  If that is a directory that exists locally
on
> every node (e.g. /tmp ), then the large scratch files do not need to go
over
> the network, which reduces wien-traffic significantly.
>
> Stefaan
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>