[Wrf-users] Radiative transfer performance in DM+SM run

[Matt Foster]

During some recent testing of a newly-configured domain (474x474, 3km) I 
noticed a rather significant difference in the performance of the 
radiative transfer (RRTM/Dudhia) when running a DM+SM build.  The 
radiative transfer timesteps are sometimes as much as 60% longer in 
DM+SM vs DM.  I've found this to be true with either OpenMPI or 
MVAPICH2, although it seems to be more pronounced with MVAPICH2.

Does anyone know why I'm seeing this behavior?

Matt

-- 
Do not go where the path may lead; go instead where there is no path and leave a trail.
-- Ralph Waldo Emerson

-------------- next part --------------
A non-text attachment was scrubbed...
Name: matthew_foster.vcf
Type: text/x-vcard
Size: 229 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/wrf-users/attachments/20110107/516791bf/attachment.vcf