AMReX-Astro / AMReX-Astro/Microphysics
Experiment with maximum order in VODE
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 43
- Forks
- 46
- Avg merge
- 2d 18h
- Merged PRs (30d)
- 15
Description
The maximum order in VODE is hardcoded to 5. We should experiment to see what happens if we use a lower limit on maximum order. On GPUs thread-local memory is limited so it is possible that the tradeoff between per-step accuracy and per-step performance is worth it when considering the overall time of the burn.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start in integration/VODE/vode_type.H at the hardcoded maximum order of 5. Experiment with a lower maximum-order limit and compare the tradeoff between per-step accuracy, per-step performance, GPU thread-local memory use, and overall burn time; done means the results support or reject lowering the limit.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- hpc, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 25/100