Welcome, Guest
Username: Password: Remember me
Forum header

TOPIC: Slow climate generation 7.4 revisited

Slow climate generation 7.4 revisited 8 years 9 months ago #460

  • Laura Rontu
  • Laura Rontu's Avatar
  • OFFLINE
  • Administrator
  • Finnish Meteorological Institute
  • Posts: 154
  • Thank you received: 8
Just would like to combine elements from three discussions: the slow climate generation discussed in summer, 7.4 problems raised by Andres and the a bit older Chilean topic. All these touch climate generation. Recently, it has been indicated that the slow climate generation is unacceptable for data assimilation experimenters, who need to quickly change from one to another large domain.

The last time I created all climate files with 7.4 and all orography and lake parameters included was this summer. For our Chilean colleagues I created all climate files of the year for this domain:

NSChile)
NLON=198
NLAT=260
NLEV=65
SOUTH=-40.90
WEST=-82.70
NORTH=-15.00
EAST=-63.00
POLAT=-90.00
POLON=0.0
NPBPTS=2
NBNDRY=8
GPHALO=8
NDTIME=600
HDFSET=regular
HIRES=0.025
;;

It did not take long time at ecgate - definetely all 12 files were made within one day. In this case, the time-independent lake and orography files were created only in the first month, which saved time for the later ones.

Iit seems that the time needed grows exponentially with increasing domain size? Perhaps it also depends on platform where the climate generation is run? Andres, did you manage to create the Estonian files in your local system? How was the time consumption there?

Re:Slow climate generation 7.4 revisited 8 years 8 months ago #498

Yes, I managed to make the files, but on local platform without flake and mso and these additional things. I did not measure time, but it was acceptable.
Time to create page: 0.074 seconds