CUDA.jl
cunumeric
CUDA.jl | cunumeric | |
---|---|---|
15 | 9 | |
1,133 | 595 | |
1.1% | 0.0% | |
9.5 | 8.5 | |
7 days ago | 3 days ago | |
Julia | Python | |
GNU General Public License v3.0 or later | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
CUDA.jl
-
Ask HN: Best way to learn GPU programming?
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.
- What's your main programming language?
-
How is Julia Performance with GPUs (for LLMs)?
See https://juliagpu.org/
-
Yann Lecun: ML would have advanced if other lang had been adopted versus Python
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.).
-
C++ is making me depressed / CUDA question
If you just want to do some numerical code that requires linear algebra and GPU, your best bet would be Julia or Python+JAX.
-
Parallélisation distribuée presque triviale d’applications GPU et CPU basées sur des Stencils avec…
GitHub - JuliaGPU/CUDA.jl: CUDA programming in Julia.
- Why Fortran is easy to learn
-
Generic GPU Kernels
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/
-
Announcing The Rust CUDA Project; An ecosystem of crates and tools for writing and executing extremely fast GPU code fully in Rust
I'm excited to eventually see something like JuliaGPU with support for multiple backends.
-
[Media] 100% Rust path tracer running on CPU, GPU (CUDA), and OptiX (for denoising) using one of my upcoming projects. There is no C/C++ code at all, the program shares a single rust crate for the core raytracer and uses rust for the viewer and renderer.
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.
cunumeric
- Announcing Chapel 1.32
-
Is Parallel Programming Hard, and, If So, What Can You Do About It? [pdf]
I am biased because this is my research area, but I have to respectfully disagree. Actor models are awful, and the only reason it's not obvious is because everything else is even more awful.
But if you look at e.g., the recent work on task-based models, you'll see that you can have literally sequential programs that parallelize automatically. No message passing, no synchronization, no data races, no deadlocks. Read your programs as if they're sequential, and you immediately understand their semantics. Some of these systems are able to scale to thousands of nodes.
An interesting example of this is cuNumeric, which allows you to take sequential Python programs that use NumPy, and by changing one line (the import statement), run automatically on clusters of GPUs. It is 100% pure awesomeness.
https://github.com/nv-legate/cunumeric
(I don't work on cuNumeric, but I do work on the runtime framework that cuNumeric uses.)
-
GPT in 60 Lines of NumPy
I know this probably isn't intended for performance, but it would be fun to run this in cuNumeric [1] and see how it scales.
[1]: https://github.com/nv-legate/cunumeric
-
Dask – a flexible library for parallel computing in Python
If you want built-in GPU support (and distributed), you should check out cuNumeric (released by NVIDIA in the last week or so). Also avoids needing to manually specify chunk sizes, like it says in a sibling comment.
https://github.com/nv-legate/cunumeric
-
Julia is the better language for extending Python
Try dask
Distribute your data and run everything as dask.delayed and then compute only at the end.
Also check out legate.numpy from Nvidia which promises to be a drop in numpy replacement that will use all your CPU cores without any tweaks on your part.
https://github.com/nv-legate/legate.numpy
-
Learning more about HPC as a python guy
Something for the HPC tools category: https://github.com/nv-legate/legate.numpy
-
Unifying the CUDA Python Ecosystem
You might be interested in Legate [1]. It supports the NumPy interface as a drop-in replacement, supports GPUs and also distributed machines. And you can see for yourself their performance results; they're not far off from hand-tuned MPI.
[1]: https://github.com/nv-legate/legate.numpy
Disclaimer: I work on the library Legate uses for distributed computing, but otherwise have no connection.
- Legate NumPy: An Aspiring Drop-In Replacement for NumPy at Scale
What are some alternatives?
LoopVectorization.jl - Macro(s) for vectorizing loops.
cupy - NumPy & SciPy for GPU
awesome-quant - A curated list of insanely awesome libraries, packages and resources for Quants (Quantitative Finance)
CudaPy - CudaPy is a runtime library that lets Python programmers access NVIDIA's CUDA parallel computation API.
cudf - cuDF - GPU DataFrame Library
numba - NumPy aware dynamic Python compiler using LLVM
Tullio.jl - ⅀
legate.pandas - An Aspiring Drop-In Replacement for Pandas at Scale
GPUCompiler.jl - Reusable compiler infrastructure for Julia GPU backends.
grcuda - Polyglot CUDA integration for the GraalVM
shared_numpy - A simple library for creating shared memory numpy arrays