[Wrf-users] Radar DFI: What should it be doing?

[Matt Foster]

I did a build of WRF 3.2.1 (DM, OpenMPI) with Radar DFI enabled.  I 
turned on TDFI and dfi_radar in the namelist file.  When I run the 
simulation (3km, 433x433 grid points), all CPUs are at 100% usage, and 
nothing is appearing to happen.  All we see in the rsl.out files is...

Ntasks in X   1, ntasks in Y   71

We are running on 72 CPUs with IO Quilting enabled.

I let it run for nearly 30 minutes, during which time there was no 
additional output in the rsl.out files, and the CPUs were at 100% for 
the entire time.  I also checked our Infiniband switch, and there was no 
activity.  To me, that indicated that nothing productive was happening.

Is there anyone on here with experience using Radar DFI that might know 
why this is happening?

One suspicion I had was that the build was also done with -DRUC_CLOUD.  
I wonder if that and Radar DFI should not be combined?

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/20110429/170bf038/attachment.vcf