Skip to content

CUDA Enabled

Latest
Latest
Compare
Choose a tag to compare
@alihakimtaskiran alihakimtaskiran released this 24 Jun 22:00
· 2 commits to main since this release
5.0.1
44bc833
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.

CUDA Enabled

"Solve Maxwell's equations faster than Maxwell"

Accelerate your CEM Workloads with CUDA

Introducing the CUDA-enabled version of glimy, now you can significantly accelerate your CEM (Computational Electromagnetics) workloads while maintaining backwards compatibility. To take advantage of this feature, follow the steps below:

  1. Install the latest versions of glimy and cupy. You can find installation instructions for cupy here. Please note that cupy is only required if you want to utilize the GPU; otherwise, you can use it in CPU-only mode.

  2. Ensure that your computer has the necessary GPU drivers and CUDA installed correctly.

To utilize the GPU for your CEM workloads, follow these steps:

  1. Call Continuum.build(gpu_flag=1) to enable GPU acceleration.

  2. Use Continuum.Render(backend="cupy") to render your CEM simulations using the cupy backend.

For a practical example of utilizing the GPU with glimy, you can refer to the following example: https://github.com/alihakimtaskiran/Glimy/blob/main/examples/7-GPU.py.

With these optimizations, you can leverage the power of CUDA and experience significantly faster CEM computations.

Assets 2