[Wrf-users] Wrf-users Digest, Vol 75, Issue 7

[jagan TNAU]

Dear Feng,

Normally the efficiency of the cluster will go down if you increase the
number of processors. However there will be improvement when you increase
the processors. If the efficiency of the 8 processors is less than two then
it is your network card. If you have InfiniBand you will get good
performance but costly affair. If you have gigabit connectivity then what
you experience will happen. Depending on budget you have you can go for 10
gig or 40 gig connectivity to improve the efficiency.


On Wed, Nov 17, 2010 at 12:30 AM, <wrf-users-request at ucar.edu> wrote:

> Send Wrf-users mailing list submissions to
>        wrf-users at ucar.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
>        http://mailman.ucar.edu/mailman/listinfo/wrf-users
> or, via email, send a message with subject or body 'help' to
>        wrf-users-request at ucar.edu
>
> You can reach the person managing the list at
>        wrf-users-owner at ucar.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Wrf-users digest..."
>
>
> Today's Topics:
>
>   1. The efficiency problem to run WRFV3.2.1 on a cluster      with 8
>      nodes (Feng Liu)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Fri, 12 Nov 2010 21:35:55 +0000
> From: Feng Liu <FLiu at azmag.gov>
> Subject: [Wrf-users] The efficiency problem to run WRFV3.2.1 on a
>        cluster with 8 nodes
> To: "wrf-users at ucar.edu" <wrf-users at ucar.edu>
> Cc: wrfhelp <wrfhelp at ucar.edu>, Feng Liu <FLiu at azmag.gov>
> Message-ID: <9BDE2A7F9712AF45A0C08451B3CD8E5C029AEF37 at mag9006>
> Content-Type: text/plain; charset="us-ascii"
>
> Hi, there,
> I'm trying to run WRF model with parallelized version with 2, 4, 8, or 16
> processors on a Linux cluster with 8 nodes (each node is formed by
> 2-quadcores). Runs got slower with increasing the number of processors (np)!
> It runs correctly on all nodes but so slow. When I switch to np=2, model is
> running on the master node only and faster. The overall time of the
> simulation is bigger than for the single node run... Is the problem
> associated with bandwidth? network card? I have no idea. Anyone have
> experienced the same problem? Thanks.
> Feng
>
>
>
> ------------------------------
>
> _______________________________________________
> Wrf-users mailing list
> Wrf-users at ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/wrf-users
>
>
> End of Wrf-users Digest, Vol 75, Issue 7
> ****************************************
>



-- 
With regards

Dr.R.Jagannathan
Professor of Agronomy,
Department of Agronomy
Tamil Nadu Agricultural University,
Coimbatore - 641 003 India

PHONE:  Mob: +91 94438 89891

DO NOT PRINT THIS E-MAIL UNLESS NECESSARY. THE ENVIRONMENT CONCERNS US ALL.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/wrf-users/attachments/20101117/e46435c2/attachment.html