Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Unifying the CUDA Python Ecosystem

developer.nvidia.com

Post reply on HN
Print view View original

11–20 of 62 posts

Re: Unifying the CUDA Python Ecosystem

#11

Post by nevi-me »

I have a RTX 2070 that's under-utilised, partly because I'm surprisingly finding it hard to understand C, C++ and CUDA by extension.

I'm self-taught, and have been using web languages and some python, before learning Rust. I hope that NVIDIA can dedicate some resources to creating high-quality bindings to the C API for Rust, even if in the next 1-2 years.

Perhaps being able to use a systems language that's been easy for me coming from TypeScript and Kotlin, could inspire me to take baby steps with CUDA, without worrying about understanding C.

I like the CUDA.jl package, and once I make time to learn Julia, I would love to try that out. From this article about the Python library, I'm still left knowing very little about "how can I parallelise this function".

Re: Unifying the CUDA Python Ecosystem

#12

Post by rrss »

andrew_v4 wrote: post #7

Just for contrast its interesting to look at an example of writing a similar kernel in Julia: https://juliagpu.gitlab.io/CUDA.jl/tutorials/introduction/ I don't think it's possible to achieve something like this in python because of how it's interpreted (but it sounds a bit like what another comment mentioned where the python was compiled to C)

I think the contrast is probably less about the language, and more about the scope and objective of the projects. the blog is describing low-level interfaces in python - probably more comparable is the old CUDAdrv.jl package (now merged into CUDA.jl): https://github.com/JuliaGPU/CUDAdrv.jl/blob/master/examples/...

here is writing a similar kernel in python with numba: https://github.com/ContinuumIO/gtc2017-numba/blob/master/4%2...

Re: Unifying the CUDA Python Ecosystem

#13

Post by jkelleyrtp »

nevi-me wrote: post #11

I have a RTX 2070 that's under-utilised, partly because I'm surprisingly finding it hard to understand C, C++ and CUDA by extension. I'm self-taught, and have been using web languages and some python, before learning Rust. I hope that NVIDIA can dedicate some resources to creating high-quality bindings to the C API for Rust, even if in the next 1-2 years. Perhaps being able to use a systems language that's been easy…

+1 Would love to see official support for CUDA for Rust.

Re: Unifying the CUDA Python Ecosystem

#14

Post by sdajk3n423 »

nevi-me wrote: post #11

I have a RTX 2070 that's under-utilised, partly because I'm surprisingly finding it hard to understand C, C++ and CUDA by extension. I'm self-taught, and have been using web languages and some python, before learning Rust. I hope that NVIDIA can dedicate some resources to creating high-quality bindings to the C API for Rust, even if in the next 1-2 years. Perhaps being able to use a systems language that's been easy…

If you are a looking to maximize use of that card, you can make about $5 a day mining crypto with the 2070.

Re: Unifying the CUDA Python Ecosystem

#16

Post by andi999 »

Actually I like pyCuda. https://documen.tician.de/pycuda/tutorial.html

You can write all the boilerplate in python and just the kernel in C (which you can pass to a string and compiler automatically in your python script). So far the workflow is much smoother than with nvcc (and creating some dll bindings for the c programm).

Re: Unifying the CUDA Python Ecosystem

#17

Post by nevi-me »

sdajk3n423 wrote: post #14
nevi-me wrote: post #11

I have a RTX 2070 that's under-utilised, partly because I'm surprisingly finding it hard to understand C, C++ and CUDA by extension. I'm self-taught, and have been using web languages and some python, before learning Rust. I hope that NVIDIA can dedicate some resources to creating high-quality bindings to the C API for Rust, even if in the next 1-2 years. Perhaps being able to use a systems language that's been easy…

If you are a looking to maximize use of that card, you can make about $5 a day mining crypto with the 2070.

No, the high electricity cost in my country + the noise pollution in the house + how much I generally earn from the machine + my views on burning the world speculatively, discourage me from mining crypto.

Perhaps my position might change in future, but for now, I'd probably rather make the GPU accessible to those open-source distributed grids that train chess engines or compute deep-space related thingies :)

Re: Unifying the CUDA Python Ecosystem

#18

Post by Tomte »

nevi-me wrote: post #11

I have a RTX 2070 that's under-utilised, partly because I'm surprisingly finding it hard to understand C, C++ and CUDA by extension. I'm self-taught, and have been using web languages and some python, before learning Rust. I hope that NVIDIA can dedicate some resources to creating high-quality bindings to the C API for Rust, even if in the next 1-2 years. Perhaps being able to use a systems language that's been easy…

> I have a RTX 2070 that's under-utilised

I've found that there are really good and beginner-friendly Blender tutorials. Both free and paid ones.

Re: Unifying the CUDA Python Ecosystem

#19

Post by SloopJon »

I thought for sure that someone would have posted a link to that xkcd comic by now. I only dabble with higher-level APIs, so I can't judge this on the merits. If NVIDIA really continues to back this, and follows through on wrapping other libraries like cuDNN, it could be a whole new level of vendor lock in as people start writing code that targets CUDA Python. I think the real test will be whether one of the big projects like PyTorch or TensorFlow gets on board.

Re: Unifying the CUDA Python Ecosystem

#20

Post by Ivoah »

SloopJon wrote: post #19

I thought for sure that someone would have posted a link to that xkcd comic by now. I only dabble with higher-level APIs, so I can't judge this on the merits. If NVIDIA really continues to back this, and follows through on wrapping other libraries like cuDNN, it could be a whole new level of vendor lock in as people start writing code that targets CUDA Python. I think the real test will be whether one of the big proj…

https://xkcd.com/927/

https://xkcd.com/1987/

Post reply on HN