Minutes of the code management group meeting 2015-03-04
From UKCA
Present
Luke Abraham (LA), Nicolas Bellouin (NB), Nick Savage (NS), Mohit Dalvi (MD), Fiona O'Connor (FOC), Alan Hewitt (AH), Mark Richardson (MR), Sandip Dhomse (SD), Graham Mann (GM), Zak Kipling (ZK)
Teleconference Numbers
- UK Freefone: 0800 917 1956
- Participant passcode: 51615813 then #
Agenda
- Minutes of the last meeing
- vn10.1
- GA7.0 update
- Rose on MONSooN
- Bugfixes
- Targets for vn10.2
- LFRic
- New supercomputer: Cray XC-40
- vn7.3 on MONSooN XC-40
- A.O.B./D.N.M.
- Date of next meeting
Minutes
Minutes of the last meeing
Actions carried forward from previous meetings
- (1) JM to check whether settings, branches in vn8.4 Release candidate are consistent with GA6/7 assessment job and also attempt tuning of SO2 scavenging parameters. In progress - liase with LA?
- carried forward - JM to chat with LA offline
- GA7.0 and SO2 still being tuned.
- New Action: JM to email LA, cc GM
- (2) CJ to circulate plots of radical concentrations in stratosphere showing odd banded structure, including version with different photolysis scheme options. Ongoing. Done a while ago - nothing much has changed?
- (2.1) FOC to put on UKESM1 document - done
- (3) CJ to email LA re problems with SO2 diagnostics for emissions. Issues related to using ASAD diagnostics package. Ongoing - needs debugging.
- (3.1) CJ to put branch on PUMA email CMG location of branch to check details - ongoing
- (4) GM check 8.4 versus 7.3 GLOMAP-mode code and report any bug fixes to Mohit. Ongoing. Still Ongoing (big task) - not just bugfixes, also functionality updates
- in progress: comparisons have been done, but reporting still needs to be done
Actions from the last meeting
- (1) UKCA Improvements for UKESM1
- (1.1): FOC to add in status and ticket numbers to each item - ongoing
- (1.2) Dry-deposition, split into 2 options: - done
- i) Quick and dirty fix (NS/FOC(?))
- ii) Future proof (AH)
- (1.3) FOC to reorder by priority - done
- (1.4) FOC/NB/GM - discuss (9) offline - (GLOMAP-mode coupling to FJX) ongoing. We could also include Apostolos Voulgarakis in this. Could this also link to the heterogenous chemistry task?
- (1.5) FOC to talk to Colin Jones about nitrate (high priority), with a offer to help Steve Rumbold - done. Progressing for vn10.2 code freeze (still at vn8.6).
- (1.6) JM to make standing telecon from January to discuss this list. Discuss with Alistair Sellar - done under UKCA coordination.
- (1.7) ALL - feed in to FOC list of priorities - ongoing (needs re-emphasising)
- CJ, AS, & FOC met after last CMG and updated document. Items need an owner if people need it raised.
- (2) vn9.2/GA7.0 update
- (2.1) JM to report back on progress at the next meeting - done
- (2.2) JM/GM to talk offline - done
- (3) Release jobs
- (3.1) LA to release RC6.2 - done: Release Job RJ4.0]
- (4) Rose stem jobs
- (4.1) MD to do for 10.2 testing (needs updates to 10.1) - done
- (4.2) LA to talk to Ros Hatcher at NCAS-CMS about NCAS plans for Rose stem - done
- Email from Ros Hatcher (2015-01-30, updated 2015-03-04)
I ran the first vn10.0 job on MONSooN a couple of days ago. So in theory you can start running 10.0 jobs now. The only thing that is missing are the prebuilds which I plan to attack today or beginning of next week. It is possible to access the shared MOSRS repositories from MONSooN. I’m currently testing out the setup that the FCM team have recommended and it appears to be working ok for me. You can now create 10.x branches on the MOSRS from MONSooN and easily use working copies in your Rose suites. You can also communicate with the mirror repositories on PUMA. It is currently slightly quicker to extract from the mirror rather than direct from the shared repository. At the moment, until the MOSRS Rose suite repository is made generally available, Rose suites need to live in the PUMA suite repository. If you want to try running a basic UM10.0 suite on MONSooN I can send you all the details for this. Let me know. As for Rose stem.... I've had nothing to do with this on MONSooN, but I believe Matthew Hambley had got this working for UM9.2. However, I'm not sure how ready for use it is. Annette & I will hopefully begin work on UM10.0 for ARCHER in earnest next week, so it will still be a little while before it is generally available. Rose stem on ARCHER has not yet been discussed. We will need it in some form or other, but not having really looked at rose-stem apart from the install process I have no idea how much work it will be. A UM release should now be available on MONSooN pretty much immediately. If I can get the responsibility for it under the UM systems team umbrella then it will become part of the Met Office standard install. This hasn't been discussed but I would assume that not all versions will be supported on ARCHER as currently, only those that are required by the community.
Action: LA to monitor this, possibly ask for a subset
- (5) Bug Reporting
- (5.1) Incorporate Bug #35 into dry deposition work (see above). - not done - Federico Centoni (FC) wants to do some sensitivities - hold off until FCs assessment
- (5.2) Bug #36 (ozone): FOC/Steven Hardiman to take on - see MOSRS ticket #102 (UKCA Bugfix for ozone column diagnostic calculation) - ongoing
- aim for vn10.2. This should be possibly be done as a bugfix package branch, as there is also a wetlands fix not on bugfixes page (Alistair Sellar found warning about wetland emissions in OffOx run)
- Action: FOC/MD to open some tickets and add them to the bugfixes page, also there are some from Action/Task 4 from Previous Meetings which need adding.
- Action: GM/MR to update bugfix page with their bug involving model_levels/wet_levels (will there even be a distinction in future, and/or could we just add a trap on this?)
- Action: NS to check with Glenn Greed about wet_levels.
- (5.3) CMG to monitor list of bugs and circulate between group for owners - ongoing
- (5.4) At CMG future meetings there should be included an appendix listing the open bugs. If a ticket is included it is not an open bug (as the owner of ticket is person doing lodging). - ongoing
- (6) Shared repository
- (6.1) ALL to migrate tickets over - ongoing
- (7) Date of next meeting
- (7.1) LA to set up doodle poll for next meeting for after the code freeze - late Feb/Early March. - done
vn10.1
- The agenda for the UM Project Board held on 2015-02-25 can be found here: https://code.metoffice.gov.uk/trac/um/wiki/ProjectDocumentation/10.x/PB_agenda_1
- Several tickets made it in to vn10.1 - see: UM Trac pages
- Ticket #102 (UKCA Bugfix for ozone column diagnostic calculation) didn't make it for vn10.1 and so has been bumped to vn10.2
- Ticket #197 has had some STASHmaster issues
- Ticket #203 had to have the science changes removed due to clashing with KGO at a late stage. Ticket #341 has been opened to add these in.
GA7.0 update
Rose on MONSooN
Bugfixes
List of open bugs
Open bugs are classed as those with no associated ticket number and/or which are not classed as lodged.
- #31 (FP exceptions in asad_sparse_vars or asad_spimjmp)
- #35 (Error in labelling: Surface resistance <--> stomatal resistance ?)
- #36 (Error in calculation of Column Ozone diagnostic)
- MOSRS ticket: #102
- Comment: To be lodged at vn10.1
- #38 (Missing logical for volcanic so2 emissions.)
- #41 (OC biomass burning emissions - incorrect conversion. Assumes emissions as kg POM, but (new) ancils are kg C)
- Met Office ticket: #6429
- Comment: To be lodged: Need to determine which ancils are POM vs C
Targets for vn10.2
- Ticket keywords - please include UKCA and SCO138
- Current vn10.2 tickets
LFRic
- Overview document on implications of GungHo! coupling for UKCA by FOC and LA:
GungHo_LFRic_UKCA_Jan2015.pdf
- Draft document by LA outlining some possible options for organisation of the UKCA code for coupling to GungHo!/LFRic:
UKCA–GungHo.pdf
- Document by NS with his thoughts on coupling UKCA to LFRic:
LFRIC_and_UKCA.pdf
New supercomputer: Cray XC-40
- Paul Cresswell has sent LA the following about Rose stem jobs on XC-40:
All the vn10.1 tests on the IBM are also available on the Cray, including Mohit's new tests, although many of them have issues - some of them are segfaulting, some have issues writing the dumps in a reasonable time (this can sometimes be fixed with the right environment variables to tweak the MPI behaviour), some of them run to completion but are not bit-comparable between runs. Mohit's new tests are in the first category. The hadgem_nd_ukca test is the only one which hits the particular error fixed by #346, so everything else must not be executing this particular piece of code (or crashing out sooner). The aqum_nd test is one of the few that seems to behave and be reproducible. Much of the UM 10.2 dev cycle will be spent fixing the others! The aqum_nd_comp_check (compiler checking) job also works, producing lots of warning messages that will be helpful in the future. Please note that no XC40 tests are available in the standard collective groups (e.g. developer, ukca), as not everyone has access to the machine yet - it would not do to have rose-stem suites failing for users who do not have accounts set up. The Cray tests are all available individually by replacing "metohpc" with "meto_xc40" in the group name. [The porting of Rose stem tests are] a joint effort - the UM System and HPC Optimisation teams are both working on it to varying degrees, but there's a lot needs doing, so if a developer/code owner/test owner wants to prioritise fixing their own job and get the fixes to us, we'll gladly accept the help. Some things will be down to the technical teams though - for example if lots of jobs are not bit comparing then that's probably an issue with the compiler settings/optimisation env vars and not something we'd expect a developer would want to focus on.