Parallelization and acceleration abstractions for code and performance portability. By abstracting away the implementation specifics, we provide a limited but unified interface for parallelization and acceleration.
- OpenMP
- MPI
- CUDA
- SYCL
This library was built for use in the 3D radiative transfer code Magritte.
This library is largely inspired by, but not as complete as:
- Grid, by Peter Boyle et al.
- Hemi, by Mark Harris et al.
- Eigen