Which Solution Methods Support GPU Acceleration?
Feko supports the use of a graphics processing unit (GPU) for simulation acceleration using the unified device architecture (CUDA) framework from NVIDIA for sequential (single CPU core) runs.
Note: Minimum
requirements for the CUDA device:
The computational phases targeted for execution on a CUDA-based GPU show a significant
speedup compared to standard CPU-based execution.- Compute capability of at least 3.5
- Driver installed on system must support CUDA 11.6.
The following solution methods support GPU acceleration:
- MoM
-
Includes SEP, VEP, planar Green's function (multilayer substrates) and thin dielectric sheets for single and double precision matrix solve (LU decomposition).
Support single or multiple GPUs.
- CMA
- Supports multiple GPUs.
- RL-GO
-
For manual (fixed) ray launching grid and flat meshes (not curvilinear).
Supports multiple GPUs.
- FDTD
- Supports single GPU.