Halide VS CUDA.jl

If CPU/GPU execution speed is the goal while simultaneously code golfing the source size, https://halide-lang.org/ might have come in handy.

This is a task where Halide https://halide-lang.org/ could really shine! It disconnects logic from scheduling (unrolling, vectorizing, tiling, caching intermediates etc), so every step the author describes in the article is a tunable in halide. halide doesn't appear to have bindings for golang so calling C++ from go might be the only viable option.

Probably would have been much easier to do 15 times in https://halide-lang.org/

The idea behind Halide is that scheduling memory access patterns is critical to performance. But, access patterns being interwoven into arithmetic algorithms makes them difficult to modify separately.

So, in Halide you specify the arithmetic and the schedule separately so you can rapidly iterate on either.

It is not the sorting per-se which was improved here, but sorting (particularly short sequences) on modern CPUs with really the complexity being on the difficulty of predicting what will work quickly on these modern CPUs.

Doing an empirical algorithm search to find which algorithms fit well on modern CPUs/memory systems is pretty common, see e.g. FFTW, ATLAS, https://halide-lang.org/

Halide https://halide-lang.org/

It doesn’t apply in this case, but in general if you really want the best vectorization I would suggest using https://halide-lang.org instead of trying to coerce your compiler.

If we drop the "APL" requirement, wouldn't Halide fit your criteria for the third?

It would also mean learning Julia, but you can write GPU kernels in Julia and then compile for NVidia CUDA, AMD ROCm or IBM oneAPI.

https://juliagpu.org/

I've written CUDA kernels and I knew nothing about it going in.

See https://juliagpu.org/

If you look at Julia open source projects you'll see that the projects tend to have a lot more contributors than the Python counterparts, even over smaller time periods. A package for defining statistical distributions has had 202 contributors (https://github.com/JuliaStats/Distributions.jl), etc. Julia Base even has had over 1,300 contributors (https://github.com/JuliaLang/julia) which is quite a lot for a core language, and that's mostly because the majority of the core is in Julia itself.

This is one of the things that was noted quite a bit at this SIAM CSE conference, that Julia development tends to have a lot more code reuse than other ecosystems like Python. For example, the various machine learning libraries like Flux.jl and Lux.jl share a lot of layer intrinsics in NNlib.jl (https://github.com/FluxML/NNlib.jl), the same GPU libraries (https://github.com/JuliaGPU/CUDA.jl), the same automatic differentiation library (https://github.com/FluxML/Zygote.jl), and of course the same JIT compiler (Julia itself). These two libraries are far enough apart that people say "Flux is to PyTorch as Lux is to JAX/flax", but while in the Python world those share almost 0 code or implementation, in the Julia world they share >90% of the core internals but have different higher levels APIs.

If one hasn't participated in this space it's a bit hard to fathom how much code reuse goes on and how that is influenced by the design of multiple dispatch. This is one of the reasons there is so much cohesion in the community since it doesn't matter if one person is an ecologist and the other is a financial engineer, you may both be contributing to the same library like Distances.jl just adding a distance function which is then used in thousands of places. With the Python ecosystem you tend to have a lot more "megapackages", PyTorch, SciPy, etc. where the barrier to entry is generally a lot higher (and sometimes requires handling the build systems, fun times). But in the Julia ecosystem you have a lot of core development happening in "small" but central libraries, like Distances.jl or Distributions.jl, which are simple enough for an undergrad to get productive in a week but is then used everywhere (Distributions.jl for example is used in every statistics package, and definitions of prior distributions for Turing.jl's probabilistic programming language, etc.).

If you just want to do some numerical code that requires linear algebra and GPU, your best bet would be Julia or Python+JAX.

GitHub - JuliaGPU/CUDA.jl: CUDA programming in Julia.

Should have (2017) in the title.

Indeed cool to program julia directly on the GPU and Julia on GPU and this has further evolved since then, see https://juliagpu.org/

I'm excited to eventually see something like JuliaGPU with support for multiple backends.

That's really cool! Have you looked at CUDA.jl for the Julia language? Maybe you could take some ideas from there. I am pretty sure it does the same thing you do here, and they support any arbitrary code with the limitations that you cannot allocate memory, I/O is disallowed, and badly-typed code(dynamic) will not compile.