[Pw_forum] Parallel version and NFS

Huiqun Zhou hqzhou at nju.edu.cn
Wed Apr 5 07:39:08 CEST 2006


Sergey,

It seems you are doing a "from scratch" SCF calculation while setting
restart_mode='restart'. If this is your case, you definitly need to set
restart_mode='from_scratch'. Using restart means you have already
a successful previous run and you want to continue the run from there
this time.

Huiqun

----- Original Message ----- 
From: <sir_puding at tut.by>
To: "Paolo Giannozzi" <pw_forum at pwscf.org>
Sent: Tuesday, April 04, 2006 10:26 PM
Subject: Re[2]: [Pw_forum] Parallel version and NFS


>>> I'm using a Linux PIII cluster with 2 proc. per node and NFS
>> don't use NFS
> I have no other possibility (I'm not a UNIX guru, and think that NFS
> is used on my cluster because '/home' file-system is the same on every
> node and all *.wfc are written to the very same  directory).
>
>>>      Starting configuration read from file     silicon.save
>>>
>>>      Failed to open file     silicon.save
>>>
>>>      Using input configuration
>
>> are you sure this is what you want? P.
>
> I'm starting calculation and  use 'restart' in order not to forget load 
> files at the
> second run. Does it influence somehow on convergence?
>
> Sergey
>
> --
> Please avoid sending me Word or PowerPoint attachments.
> See http://www.gnu.org/philosophy/no-word-attachments.html
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
> 




More information about the Pw_forum mailing list