Bringing Together the GPU Computing Ecosystem for Python

John MelonakosAnnouncements, ArrayFire, Computing Trends, CUDA, Open Source, Python Leave a Comment

To date, we have not done a lot for the Python ecosystem. A few months ago, we decided it was time to change that. Like NVIDIA said in this post, the current slate of GPU tools available to Python developers is scattered. With some attention to community building, perhaps we can build something better — together.

NVIDIA spoke some about its plans to help cleanup the ecosystem. We’re onboard with that mentality and have two ways we propose to contribute:

  1. We’re working on a survey paper that assesses the state of the ecosystem. What technical computing things can you do with each package? What benchmarks result from the packages on real Python user code? What plans does each group have in the near term to build things for Python users?
  2. We’ll do the work to gather together folks into regular community roundtable Zoom calls where we can inform each other about our work, collaborate as much as we can to avoid duplication of effort, and ultimately produce something very nice for Python users.

At ArrayFire, we’ve been building interpreted GPU computing tools for 14 years and look forward to sharing the wisdom we have with this corner of the market. With a little attention to collaboration, I imagine that we can really deliver something wonderful to Python users, that minimizes the need to change and code and maximizes performance.

To be included in the mailing list and to be informed about the roundtable calls, shoot me an email directly: john@arrayfire.com

I’m personally doing a lot of the work in this space as a Q2 objective of mine. I’m a fan of “build-in-public” initiatives, so if you want to follow my day-to-day on this, follow me on Twitter too, @melonakos!

Leave a Reply

Your email address will not be published. Required fields are marked *