05-09-2024 Edinburgh Meeting Notes
- Deepak Tunuguntla
Analytics
Date
05-09-2024
Discussion topics
Item | Presenter | Notes |
|---|---|---|
Benchmarking problems for DEM (where Hongyang is present) | Daniel & Retief | Definition of benchmark problems: same as our academic reference problems? Has it been verified (algorithmically and theoretically), has it been compared and validated (against experiments ). Benchmarking: pure performance measure against different methods. Difference between single core, multi-core, GPU/CPU, and for ML different architectures, and for UP different models (mesoparticle to begin with). |
Contact model calibration | JP & Roxana | Contact models, efficiency of different calibration & optimization strategies, does you calibration depend on regimes (solid and fluid) |
Online database session (where Hongyang is present) | JP Morrissey | Possibility to add new contact models, a GUI on GrainDB. Contact models that can be plugged into various codes, e.g., Python class ↔︎ Jason. Can we do something similar with our upscaling approaches? For example, a postprocessing tool to identification of time scales and length scale from existing data |
Action items
Decisions
- Two tangible tasks defined