fbpic VS codon

When I wrote my bachelor thesis years back I worked on a particle-in-cell code [1] that makes heavy use of numba for GPU kernels. At the time it was the most convenient way to do that from python. I remember spending weeks to optimizing these kernels to eek out every last bit of performance I could (which interestingly enough did eventually involve using atomic operations and introducing a lot of variables[2] instead of using arrays everywhere to keep things in registers instead of slower caches).

I remember the team being really responsive to feature requests back then and I had a lot of fun working with it. IIRC compared to using numpy we managed to get speedups of up to 60x for the most critical pieces of code.

[1]: https://github.com/fbpic/fbpic

We used numba to accelerate the code and most importantly write GPU kernels for the heavy parts. I remember spending hours optimising my code to eek out the most performance possible (which eventually meant using atomics and manually unrolling many loops because somehow this was giving us the best performance) but honestly I was really happy that I didn't need to write cuda kernels in C and generally it was pretty easy to work with. I remember back then the documentation was sometimes a little rough around the edges but the numba team was incredibly helpful and responsive. Overall I had a great time.

[0] https://github.com/fbpic/fbpic

https://github.com/exaloop/codon/blob/develop/LICENSE

Here are some others: https://github.com/search?q=%22Business+Source+License%22+%2...

I found at least one project that managed to compile python AOT to LLVM https://github.com/exaloop/codon. Even if LLVM is more expressive than Dart Kernel, that should at least be some evidence that this might not be too impractical.

Their fannkuch benchmark seems to be a bit dishonest. They claim an enormous perf delta on https://exaloop.io/benchmarks.html but fannkuch uses factorial a lot and they define factorial with a very small (n=20) table: https://github.com/exaloop/codon/blob/fb461371613049539654c1...

Disclaimer: I've worked on several Python runtimes and compilers, but I'm not by any means out to get Codon. Just happened across this by accident while looking at their inline LLVM, which is neat.

If you literally want Python, but compiled ... Look at Codon: https://github.com/exaloop/codon

Another "Python with high-performance compiled builds" would be https://github.com/exaloop/codon.

This is the project being discussed: https://github.com/exaloop/codon

Try Codon? https://github.com/exaloop/codon

https://github.com/exaloop/codon might save you from the rewrite.