[mpas-developers] update on date/time manager
Michael Duda
duda at ucar.edu
Thu Jun 9 10:32:36 MDT 2011
Phil,
thanks for the insight on the additional burden in adopting the
full ESMF library. I'd also favor using just the stand-alone ESMF
time manager, even if it does require us to write a bit more of
our own code within the MPAS time manager wrapper.
I'll have a go at adding the ability to set time intervals in
years and months via the ESMF 2.1.1 code. Thanks, again!
Cheers,
Michael
On Wed, Jun 08, 2011 at 04:56:06PM -0600, Jones, Philip W wrote:
>
> Michael,
>
> I don't have recent experience with their build, but suspect building is not a real problem. However...
>
> ESMF is a very large library and will require an ESMF_init somewhere in your driver or init code. Used to be that ESMF_init also did an MPI_init (with a serial wrapper if you're not building parallel code) - not sure whether that's still the case. The point is that it will affect code beyond just what you want to use.
>
> Guess I have a bit of a bias against ESMF for MPAS at this time, just because I think you want to be a bit more nimble for now. It might make sense later. It's also one of my pet peeves - frameworks that force you to adopt the whole thing instead of being able to use/adopt in a piecewise manner.
>
> You also wrote:
> ... it does not support getting
> intervals in terms of years and months (though it does, contrary
> to what I had thought, permit the setting of intervals in years
> and months).
>
> Actually, this would be fine. We want to set monthly, yearly intervals, but when getting the interval, we usually want it back in seconds, not MM or YYYY.
>
> Phil
>
> ---
> Correspondence/TSPA/DUSA EARTH
> ------------------------------------------------------------
> Philip Jones pwjones at lanl.gov
> Climate, Ocean and Sea Ice Modeling
> Los Alamos National Laboratory
> T-3 MS B216 Ph: 505-500-2699
> PO Box 1663 Fax: 505-665-5926
> Los Alamos, NM 87545-1663
>
>
>
More information about the mpas-developers
mailing list