Wavesim Python Release
October 2024: We released a Python implementation of Wavesim v0.1.0 - alpha for solving the Helmholtz equation through domain decomposition. With this new framework, we simulated a complex 3D structure of a remarkable 315×315×315 wavelengths (3.1⋅10^7) in size in just 379 seconds by solving over two GPUs, 1.93 times the size of the largest simulation on a single GPU without domain decomposition.
Matlab Software Update
March 2024: A new version of Wavesim was released on February 27 on GitHub. The main feature of the new version is the use of a CUDA-based acceleration module (cumex) that can provide speed-up optimisation of up to x2 times.
WaveSim
World’s fastest full Maxwell & Helmholtz solver by Ivo Vellekoop
06/03/2024: You can download the presentation slides from the Wavesim Seminar organised by the Institute of Electronic Structure and Laser (IESL) at the Foundation for Research and Technology-Hellas (FORTH)
Publications
If you use Wavesim in your work, please cite the below publications.
The main reference for Wavesim, introducing the modified Born series approach for the Helmholtz equation, which was later generalized to vector fields.
-
Osnabrugge, G., Leedumrongwatthanakun, S., & Vellekoop, I. M. (2016). A convergent Born series for solving the inhomogeneous Helmholtz equation in arbitrarily large media. Journal of Computational Physics, 322, 113-124. https://doi.org/10.1016/j.jcp.2016.06.034​​
Introduced a modified fast Fourier transform that enabled having an ultra-thin boundary layer, significantly reducing the computational costs and memory requirements.​​​
-
Osnabrugge, G., Benedictus, M., & Vellekoop, I. M. (2021). Ultra-thin boundary layer for high-accuracy simulations of light propagation. Optics Express, 29(2), 1649-1658. https://doi.org/10.1364/OE.412833
​
Introduced the domain decomposition of the modified Born series for the Helmholtz equation and the Python implementation of Wavesim.
-
mache, S., & Vellekoop, I. M. (2024). Domain decomposition of the modified born series approach for large-scale wave propagation simulations. arXiv preprint. arXiv:2410.02395.​​
​​
​This page includes a list of publications that have cited the above.
WAVESIM ROADMAP
Stay tuned...