Bohrium provides a runtime environment for efficiently executing vectorized applications using your favorite programming language Python/NumPy, C#, F# on Linux, Windows and MacOSX.
Forget handcrafting CUDA/OpenCL to utilize your GPU, forget threading, mutexes and locks to utilize your multi-core CPU and forget about MPI to program your cluster just Bohrium!
|Multi-Core CPU||Many-Core GPU||Python2/NumPy||Python3/NumPy||C++||.NET|
- Lazy Evaluation, Bohrium will lazy evaluate all Python/NumPy operations until it encounters a “Python Read” such a printing an array or having a if-statement testing the value of an array.
- Views Bohrium supports NumPy views fully thus operating on array slices does not involve data copying.
- Loop Fusion, Bohrium uses a fusion algorithm that fuses (or merges) array operations into the same computation kernel that are then JIT-compiled and executed. However, Bohrium can only fuse operations that have some common sized dimension and no horizontal data conflicts.
- Lazy CPU/GPU Communiction, Bohrium only move data between the host and the GPU when the data is accessed directly by Python or a Python C-extension.
- Please note:
- Bohrium is a 64bit project exclusively.
- We are working on a Windows version that uses the .NET frontend and CPU backend.