MetOffice / MetOffice/lfric_core
Clock to support sub-stepping
- Dominant language
- Fortran
- Stars
- 26
- Forks
- 73
- Avg merge
- 1d 8h
- Merged PRs (30d)
- 4
Description
Some algorithms need to iterate within a timestep. This is currently handled with a simple loop within that algorithm. Unfortunately this leads to a lot of plumbing in the algorithm to support it. e.g. tracking the length of a substep, i.e. `dt_substep`.
A proper solution is needed.
Contributor guide
No contributing guide indexed for this repository
Research direction
Start by examining the clock implementation and the algorithms that currently use simple loops for sub-stepping. Identify how timestep length and substep state are currently passed through the algorithm, then define what a clock-level solution would need to provide. Done should mean algorithms can iterate within a timestep without extensive substep plumbing.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- fortran
- Domain
- infrastructure
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100