GPUODEBenchmarks
DiffEqGPU.jl
| GPUODEBenchmarks | DiffEqGPU.jl | |
|---|---|---|
| 3 | 2 | |
| 24 | 295 | |
| - | 3.4% | |
| 6.7 | 5.3 | |
| about 1 year ago | about 1 month ago | |
| Cuda | Julia | |
| MIT License | MIT License |
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.
GPUODEBenchmarks
DiffEqGPU.jl
-
2023 was the year that GPUs stood still
Indeed, and this year we created a system for compiling ODE code not just optimized CUDA kernels but also OneAPI kernels, AMD GPU kernels, and Metal. Peer reviewed version is here (https://www.sciencedirect.com/science/article/abs/pii/S00457...), open access is here (https://arxiv.org/abs/2304.06835), and the open source code is at https://github.com/SciML/DiffEqGPU.jl. The key that the paper describes is that in this case kernel generation is about 20x-100x faster than PyTorch and Jax (see the Jax compilation in multiple ways in this notebook https://colab.research.google.com/drive/1d7G-O5JX31lHbg7jTzz..., extra overhead though from calling Julia from Python but still shows a 10x).
The point really is that while deep learning libraries are amazing, at the end of the day they are DSL and really pull towards one specific way of computing and parallelization. It turns out that way of parallelizing is good for deep learning, but not for all things you may want to accelerate. Sometimes (i.e. cases that aren't dominated by large linear algebra) building problem-specific kernels is a major win, and it's over-extrapolating to see ML frameworks do well with GPUs and think that's the only thing that's required. There are many ways to parallelize a code, ML libraries hardcode a very specific way, and it's good for what they are used for but not every problem that can arise.
-
Julia GPU-based ODE solver 20x-100x faster than those in Jax and PyTorch
Link to GitHub repo from the abstract: https://github.com/SciML/DiffEqGPU.jl
What are some alternatives?
jax - Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
DiffEqBase.jl - The lightweight Base library for shared types and functionality for defining differential equation and scientific machine learning (SciML) problems
hn-search - Hacker News Search
SciMLSensitivity.jl - A component of the DiffEq ecosystem for enabling sensitivity analysis for scientific machine learning (SciML). Optimize-then-discretize, discretize-then-optimize, adjoint methods, and more for ODEs, SDEs, DDEs, DAEs, etc.