A Look at Chapel, D, and Julia Using Kernel Matrix Calculations

Re:A Look at Chapel, D, and Julia Using Kernel Matrix Calcul... by eigenspace on Wednesday June 03, 2020 @02:08PM (#23406625)

This was a nice writeup!

I know there was an explicit mention that non-standard library packages were omitted due to fiddlyness in installation, but with Julia using the manifest system, providing reproducible setups should be easy, and for things like this, Chris Elrod's LoopVectorization.jl [1] should offer substantial speedups with some simple macro annotations.

[1] https://github.com/chriselrod/LoopVectorization.jl

[ Parent | on HN ]

Re:A Look at Chapel, D, and Julia Using Kernel Matrix Calcul... by cbkeller on Wednesday June 03, 2020 @04:48PM (#23408512)

Chris deserves some sort of prize for that package. Delivered some factor of 2 or greater speedups for significant portions of my code with just an `@avx`

[ Parent | on HN ]