[Liwg-core] JG-BG Morning Updates

Katherine Thayer-Calder katec at ucar.edu
Tue Dec 11 09:02:46 MST 2018


Hi all,

I thought I would send out morning updates each day as long as the spin-up
is going, to keep everybody posted about progress or any issues we are
having. However, this will be a lot of emails that the whole liwg-core may
not want to have. So, if you would like daily updates on the JG-BG spinup,
please reply to me, and I'll start a new list for this purpose.

The run is moving along. We have 9 years of data in the archive here:
/gpfs/fs1/scratch/katec/CESM21-CISM2-JG-BG-Dec2018/archive/BG_iteration_1
We are getting about 17.5 simulated years/real day, or 7 years in a 9.5
hour run. The runs are set to automatically resubmit after each 7 year
segment until we get to 35 years.

This morning, the next 7 year segment started running at about 5:30a, and
when I got into the office at 8:30a, I found out that it had hung (no more
processing, but the job was continuing on) soon after it launched this
morning with a Cheyenne-related MPT error. I killed the job and resubmitted
it, and emailed CISL asking for them to refund those hours of lost compute
time to our project. They have done this for us in the past. The current
job is progressing without issues. It's been about 25 minutes and we are
more than a month into the next 7 year segment (0010-02-08).

Ok, that's the update for today, Tuesday December 11. Again, please email
me if you wish to continue getting updates.

Thanks!
Kate Thayer-Calder
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cgd.ucar.edu/pipermail/liwg-core/attachments/20181211/39bddacb/attachment-0001.html>


More information about the Liwg-core mailing list