News for hackers. Stuff that matters.

A comment on Provably space-efficient parallel functional programming.

Re:Provably space-efficient parallel functional programming by joe_the_user on Thursday January 13, 2022 @06:16PM (#29928370)
Looks exciting,

Quick question. What sort of parallelism are they discussing, Mimd or Simd?

Notably, can what they're discussing run on a GPU?

Difference here: https://en.wikipedia.org/wiki/Flynn%27s_taxonomy

[ Parent | on HN ]
Re:Provably space-efficient parallel functional programming by shwestrick on Friday January 14, 2022 @10:42AM (#29935801)
A combination of both. This work targets multicore, but execution on the GPU is definitely possible. It's just a different line of research. I'd highly recommend taking a look at Futhark (https://futhark-lang.org/), which has a similar programming model (purely functional and parallel), but targets GPU. The compilation strategy is quite a bit different!
[ Parent | on HN ]
Re:Provably space-efficient parallel functional programming by joe_the_user on Friday January 14, 2022 @06:07PM (#29941753)
I have looked Futhark, however...

What I find more interesting is things like Dyna-Soar[1] as well as the parent project/paper. Futhark seems primarily very tensor-type computation specific (like APL+ extra features) and I'm interested in things that more naturally prone to code and data divergence and in trying to find ways to compensate for this.

[1] https://2019.ecoop.org/details/ecoop-2019-papers/12/DynaSOAr...

[ Parent | on HN ]