...
Plan for MercuryMay (draft)
- Plan is to spend a month and implement/improve/document features we deem important.
- This can be topics particular to your research, but more improtantly general improvements.
- We make a list of topics, everyone picks what he will work on (in groups, possibly).
- We have daily meetings to discuss progress.
- At the end of the month we release a new version of MercuryDPM.
@all: please add topics to this list; ideally before next weeks meeting; who works on what can be decided later
| Topic | Subtopic | Who takes lead? | Additional notes | |
|---|---|---|---|---|
| Finish JIRA Jobs | All | Everyone who still has open JIRA jobs, finish them | ||
| Coupling with oomph-lib | ViPr team and Mitchel | Define, and start to implement, an interface between MercuryDPM and oomphlib | ||
| Documentation | Force laws | Thomas | Add references, explanations to the force laws | |
| ... | A lot of code needs documentation; this is a good place to start for beginners to get familiar with the code | |||
| Intro to svn, cmake, cluster use | Anthony | At the begin of MercuryMonth, the new members need an introduction into teh tools | ||
| Efficiency | FLTO | Thomas | Inlining doesn't work across object files; using link-time optimization should improve that | |
| dynamic casts | Thomas | A big speed hit is the use of dynamic casts; so they should be removed | ||
| non-spherics | There are many unnecessary and costly string comparisons in the non-spherical code | |||
| break and continue | Thomas | Using break and continue, and unneccesary if's slow down your code | ||
Moved plan to separate webpage: MercuryMay 2019