Home - orlandoacevedo/MCGPU GitHub Wiki
TODO
Critical Path (NSF Objectives)
Flexibility Finish by December 2016
↙ ↘
Solutes Step Parallelism Finish by May 2017
↓
QM/MM Finish by August 2017
Chemistry/Major Functionality
- Support molecule flexibility (In progress: Brandon Su16, Mitch Fa16)
- Implement/find code for NLC on CPU
- Integrate Verlet on CPU (from branch)
- Parallelize NLC on GPU
- Parallelize Verlet on GPU
- Support NPT ensembles
- Quadratic feathering
- Ewald summations
- Step parallelism
- Free energy perturbation calculations
- Solutes
- QM/MM
Design, Cleanup, and Documentation
- Update Makefile to compile with GCC without OpenACC (CPU-only/no GPU)
- Add tests from Orlando's 5/26 e-mail
- Refactor Simulation/SimBox/etc. design per 5/26 discussion
- Fix CLI args to accommodate varying cutoff methods (brute-force, matrix, NLC, Verlet)
- Add testing capabilities for new calculation strategies
- Update Makefile to compile with PGCC with -ta=multicore (In progress: Brandon)
- Update Makefile to compile with GCC with OpenACC
Inessential
- Zmat explicit conversion script - add implicit angles to Z-matrix file