DARPA looks to propel parallelism with PAPPA


PAPPA specifically seeks to fill the gap between programing frameworks that automate parallelism but lack the ability to scale and popular data science programming languages like Python and R that nevertheless have shown little ability to scale on massively parallel systems.

Cough. JULIA. Cough.
Seriously though - should Julia Computing put in an application?

Also the 10000x speedup rather ignores those pesky serial parts which can’t be speeded up.


I work on the SocialSim project. Some of the code for the next challenge (December) will likely be Julia so I will let the new P.I. know about it. :slight_smile:


Yeah, this seems like an obvious way for Julia Computing to fund some advanced compiler development with DARPA money, and show everyone Julia’s abilities in this space. With the budget DARPA notes they could just hire some full-time engineers to work exclusively on the project, and Julia’s mostly already pretty excellent at working on this kind of project.