Internal:PDM ND: Difference between revisions
Jump to navigation
Jump to search
Spyridon97 (talk | contribs) |
Spyridon97 (talk | contribs) |
||
| Line 13: | Line 13: | ||
==Summer 2020== | ==Summer 2020== | ||
# Clean-up of PDT code-base | # Clean-up of PDT code-base | ||
# | # Use modern C++20 (whatever is available for the time being) | ||
# | # Replace boost:threads with std::thread / std::jthread | ||
# | # Replace CGAL with PCK predicates | ||
## Paper: [https://hal.inria.fr/hal-01225202 Predicate Construction Kit Paper] | ## Paper: [https://hal.inria.fr/hal-01225202 Predicate Construction Kit Paper] | ||
## Code: [https://gforge.inria.fr/frs/?group_id=5833 Predicate Construction Kit Code] | ## Code: [https://gforge.inria.fr/frs/?group_id=5833 Predicate Construction Kit Code] | ||
# | # Select an optimal Cell-location algorithm | ||
# | # Use of Memory Pools | ||
# Add Memory Limit | |||
# Replace Master-worker parallelism with speculative parallelism | # Replace Master-worker parallelism with speculative parallelism | ||
# | # Implement Rupert's Algorithm | ||
# | # Remove slivers using PODM 3D/4D Heuristic method | ||
# | # Create Sizing Function pointer for adaptive mesh generation | ||
# Template for 2-3-4 Dimensions | # Template for 2-3-4 Dimensions using PODM4D Triangulation | ||
# Weighted Delaunay | # Weighted Delaunay | ||
# Addition of PDR for distributed computations | # Addition of PDR for distributed computations | ||
# Decoupling of Geometry, Parallelism, and Load-Balancing with the usage of PREMA | |||
# Add Constrained Delaunay Triangulation | |||
# Handle inputs of Images | |||
# Conform data-structures and algorithm for [https://www.scorec.rpi.edu/~shephard/FEP19/papers/Dan-Ibanez-Thesis.pdf CONFORMAL MESH ADAPTATION ON HETEROGENEOUS SUPERCOMPUTERS] | # Conform data-structures and algorithm for [https://www.scorec.rpi.edu/~shephard/FEP19/papers/Dan-Ibanez-Thesis.pdf CONFORMAL MESH ADAPTATION ON HETEROGENEOUS SUPERCOMPUTERS] | ||
# Distinguish basic meshing operations that could be changed/defined based on needs, e.g. Delaunay property could be changed to advancing-front property. | # Distinguish basic meshing operations that could be changed/defined based on needs, e.g. Delaunay property could be changed to advancing-front property. | ||
Revision as of 08:17, 25 June 2020
Parallel Delaunay Mesher ND
Documentation
Summer 2020
- Read Dynamic Parallel 3D Delaunay Triangulation (PDT) ✔
- Read PODM3D ✔
- Read PODM4D ✔
- Read Q-HULL ND Delaunay Triangulator (TODO)
Requirements
Summer 2020
- Clean-up of PDT code-base
- Use modern C++20 (whatever is available for the time being)
- Replace boost:threads with std::thread / std::jthread
- Replace CGAL with PCK predicates
- Select an optimal Cell-location algorithm
- Use of Memory Pools
- Add Memory Limit
- Replace Master-worker parallelism with speculative parallelism
- Implement Rupert's Algorithm
- Remove slivers using PODM 3D/4D Heuristic method
- Create Sizing Function pointer for adaptive mesh generation
- Template for 2-3-4 Dimensions using PODM4D Triangulation
- Weighted Delaunay
- Addition of PDR for distributed computations
- Decoupling of Geometry, Parallelism, and Load-Balancing with the usage of PREMA
- Add Constrained Delaunay Triangulation
- Handle inputs of Images
- Conform data-structures and algorithm for CONFORMAL MESH ADAPTATION ON HETEROGENEOUS SUPERCOMPUTERS
- Distinguish basic meshing operations that could be changed/defined based on needs, e.g. Delaunay property could be changed to advancing-front property.
- Insert Operations
- Insert Point
- Insert Constrained Point
- Insert Point based on Image
- Remove Operations
- Remove Point
- Remove Constrained Point
- Remove Point based on Image
- Define rules for Mesh Generation
- Insert Operations