[HN Gopher] TORAX is a differentiable tokamak core transport sim...
       ___________________________________________________________________
        
       TORAX is a differentiable tokamak core transport simulator
        
       Author : yeldarb
       Score  : 31 points
       Date   : 2024-05-01 15:47 UTC (2 days ago)
        
 (HTM) web link (github.com)
 (TXT) w3m dump (github.com)
        
       | yeldarb wrote:
       | Found this really cool; I didn't even know Deepmind was working
       | on Fusion research https://www.wired.com/story/deepmind-ai-
       | nuclear-fusion/
        
         | soggybread wrote:
         | Here's an archive.org link:
         | https://web.archive.org/web/20220217012159/https://www.wired...
        
       | aqme28 wrote:
       | Very interesting that it's coming from Google. I did my masters
       | in tokamak simulation, so my first question is about performance.
       | Python is very rarely used in this space just for performance
       | reasons. Even though Python can call out to BLAS or whatever,
       | it's still usually worth it to code in Fortran or C or maybe
       | Julia.
        
         | cokernel_hacker wrote:
         | This python actually builds a graph under the hood which then
         | gets JIT compiled for CPU/GPU/TPU.
        
           | senseiV wrote:
           | Does a TPU have XLA-graph for GPUs Cuda-graphs? Not sure on
           | TPU theory
        
         | uoaei wrote:
         | It's built on JAX, not vanilla Python.
         | 
         | The metric being optimized is not just performance, but also
         | the ability to build reasonably performant workflows with
         | arbitrary differentiable (i.e., ML) inputs and outputs.
        
         | nestorD wrote:
         | I am doing quite a bit of work with JAX (the Python library
         | used here) in a high-performance numerical computing context.
         | 
         | On GPU/TPU, it is not going to reach perfect 100% hardware
         | usage, but it is going to get close enough (far above vanilla
         | Python performance) and be significantly more productive than
         | alternatives.
         | 
         | That makes it a sweet spot for research (where you will want to
         | tweak things as you go) and extremely complex codes (where you
         | already need to put your full focus on the correctness of the
         | code). I highly recommend it to domain experts who need
         | performance for their research project.
        
       | TaylorAlexander wrote:
       | I just noticed this podcast episode on Deep RL for fusion
       | reactors was recently published, if anyone likes this stuff. I
       | have not listened yet, but this podcast in general is great.
       | 
       | https://twimlai.com/podcast/twimlai/controlling-fusion-react...
        
       ___________________________________________________________________
       (page generated 2024-05-03 23:00 UTC)