@giuseppebilotta@jay SYCL and OpenCL are central to this effort, neither are on their own are enough to be competitive on every front with the CUDA ecosystem.
This lecture focuses on the following topics:
✅ Optimized Matrix Multiplication
✅ Shared Memory Techniques for CUDA
✅ Implementing Shared Memory Optimization
✅ Translating Python to CUDA and Performance Considerations
✅ Numba: Bringing Python and CUDA Together
#HPC#CUDA#OpenCL#LAPACK
If you had to do a lot of linear least square solves, with potentially rank-deficient matrices, what would you use on a GPU? On CPUs, LAPACK's DGELSY does work, but most GPU libraries seem to not implement routines for rank-deficient matrices.
A new crash course for getting started with #CUDA with #Python by Jeremy Howard 🚀. CUDA is NVIDIA's programming model for parallel computing on GPUs. CUDE is being used by tools such as #PyTorch#tensorflow and other #deeplearning and LLMs frameworks to speed up calculations. The course covers the following topics:
✅ Setting up CUDA
✅ CUDA foundation
✅ Working with Kernel
✅ CUDA with PyTorch
@VileLasagna Has a blog post on the relative speed of different #GPU compute frameworks on the same hardware and driver.
Tl;dr: on an #Nvidia card, with Nvidia drivers, #CUDA is the slowest, by far. Fastest is our old stalwart #OpenCL - almost twice as fast when used only for compute. #Vulcan is good, and the least affected by using the card for your desktop at the same time. Read it - it's good.
@hyc@VileLasagna
It's a simple computing test. Results are of course going to differ by the task - and ultimately the only benchmark that matters is your production code.
With that said, there's nothing odd going on with the source that I can see; this kind of simple structure isn't very rare in "real" code; and the difference is quite striking.
It's not the only data point I've seen - there's a recent CFD code that also shows impressive results with opencl.
I've been facing many issues with using #Poetry (#pythonpoetry) with my #Python based #objectdetection project. I love Poetry for publishing packages, but think that #conda would be better since I have to deal with #CUDA and whatnot. Anyone familiar with a way to use pyproject.toml for publishing and building packages, even if Poetry isn't being used for dependency management?
@kytta Interesting - what do you recommend I migrate to, especially if I want to use conda? I imagine if using any build tool with conda, the build tool would not be able to package the program if dependencies are installed with conda and not directly with pip - correct?
@slashtechno yeah, that's where my expertise ends, as I do not know anything about conda. In Python packaging, you don't package the dependencies with the app, so you don't need to install them. Instead, the wheels just have the package names, so pip can resolve the dependency tree and pull the needed packages by itself