Can you use a Raspberry Pi as (local) CPU core with the Parallel Computing Toolbox and/or Distributed Computing Server?

6 views (last 30 days)
I would like to speed up computations using (possibly multiple) Raspberry Pis along with the Matlab Parallel Computing Toolbox and/or Matlab's Distributed Computation Server Toolbox. Is it possible to connect a Raspberry Pi to my machine and handle it kind of similar to a local core? This way, I wouldn't need to install Matlab on the Pi (which I think isn't possible) but could make use of the Raspberry Pi's CPU. Or is it maybe possible to build a Raspberry Pi cluster with the Distributed Computation Server Toolbox?
  1 Comment
Nicolas Zink
Nicolas Zink on 9 Dec 2019
I am having the exact same question. Since the original question was asked in 2016 already, is there anyone who has tried to set up Rasberry Pis as Matlab workers for parallel computing?

Sign in to comment.

Answers (1)

Malcolm Lidierth
Malcolm Lidierth on 18 Jun 2016
You can run m-code using Octave on a Pi if that helps. Your bottleneck will likely be data exchange (as e.g. when using a GPU).

Categories

Find more on MATLAB Support Package for Raspberry Pi Hardware in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!