[Wrf-users] machine to run WRF

Abdullah Kahraman havadurumu at gmail.com
Tue Jan 14 10:45:15 MST 2014


Theoretically, a faster HDD would help with writing output. But if the
history intervals you'll use are not extremely tight (i.e. in order of
minutes or seconds), I don't think that effects the performance of WRF very
much. However, not much for running the model, but during postprocessing, a
faster reading can become quite important.

RAM size would be crucial depending on your domain size (in terms of dx, dy
and number of vertical levels). I guess a wiser choice would be increasing
ram, if you are dealing with very large domains, rather than HDD. The RAM
transfer rate (and that of mainboard) should be high as well, since maybe
almost all performance depends on the CPU, RAM and the transfers between
them.

If there is no problem with the domain sizes or excessive postprocessing, I
would improve the CPUs instead of any other hardware components. But maybe
searching for benchmarks online would help better with your decision.


*Abdullah KAHRAMAN*
Visiting Research Scholar
Department of Meteorology
Pennsylvania State University
410 Walker Building
University Park, PA 16802
*Meteogreen: http://web.itu.edu.tr/kahramanab/index.php
<http://web.itu.edu.tr/kahramanab/index.php>*


On Mon, Jan 13, 2014 at 4:07 PM, Julio Castro <julio.castro at arcadis.cl>wrote:

>  Hi all,
>
>
>
> We are getting a dedicated machine to run WRF. 32 GB RAM and 2 *Intel
> Xeon E5-2450 2.10GHz, 20M Cache (8 cores each) * with some  *7.2K RPM
> SATA 3Gbps HDs, my question is if there is any point in getting faster HD,
> say 10 K or 15K RPM instead of the 7.2K RPM?*
>
>
>
> Thanks in advance
>
>
>
>
>
> *Julio Castro M.* | Especialista Senior en Calidad de Aire |
> julio.castro at arcadis.cl
>
> ARCADIS Chile | Antonio Varas #621| Providencia
>
> CP 7500655 | Santiago | Chile
>
> T. + 56 2 2 3816248 | F. + 56 2 2 3816001
>
> www.arcadis.cl
>
>
>
> ARCADIS, Imagine the result
>
> *Antes de imprimir este correo piense bien si es necesario. ARCADIS por un
> mundo sustentable.*
>
>
>
> ------------------------------
> Este correo electrónico es confidencial. Está destinado para su uso sólo
> por parte del destinatario.
> Si usted no es el destinatario previsto, le solicitamos notificarlo al
> emisor y borrar este correo electrónico, y cualquier otro adjunto(s), sin
> copiarlo, reenviarlo, divulgarlo o usarlo de cualquier otra forma.
>
> This e-mail is confidential and may also be privileged. It is intended for
> use by the addressee only.
> If you are not the intended addressee, we request that you notify us
> immediately and delete this e-mail, and any attachment(s), without copying,
> forwarding, disclosing or using it in any other way.
>
> _______________________________________________
> Wrf-users mailing list
> Wrf-users at ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/wrf-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/wrf-users/attachments/20140114/af5007d5/attachment.html 


More information about the Wrf-users mailing list