[GRASS-user] hardware requirements for running r.sun

[Dylan Beaudette]

On Tuesday 30 June 2009, Hamish wrote:
> buraq wrote:
> > I have a notebook with T7200 Intel CPU and r.sun completes
> > one day's calculation between 3-4 mins(with linke turbidity raster
> > and shadowing, step = 1). I want to run r.sun with step=0.5 in a
> > large scale and detailed area.
> > So, I need to reduce the calculation time for calculating 365 day data.
> > Could you give me advices about hardware for reducing
> > calculation time?
>
> see tips here:
>   http://grass.osgeo.org/wiki/r.sun
>
> (run r.horizon first)
>
> 24 hours to complete isn't so bad :)
> suggestion: to take advantage of dual processor split up operations and
> (if running linux) run two concurrent mapsets for day 0-180 and 181-365.
> then when done use g.mapsets to add second mapset to first's search path.
>
> wish: someone in the know adds OpenMP* support to r.sun
> [*] (or pthreads, but OpenMP seems easier)
>      http://grass.osgeo.org/wiki/OpenMP
>
>
> Hamish
>

Not trying to honk one's own horn, but an application of r.sun and time 
requirements was recently published here:

http://soil.scijournals.org/cgi/content/abstract/73/4/1345

Let me know if you want a PDF

Dylan




-- 
Dylan Beaudette
Soil Resource Laboratory
http://casoilresource.lawr.ucdavis.edu/
University of California at Davis
530.754.7341