Clojure is Not Afraid of the GPU - Dragan Djuric

Channel:ClojureTV · Length:42:13 · Published:2016-11-17 · Views:6883

Watch on YouTube

Summary

Dragan Djuric presents Clojure libraries for GPU and numerical computing, including ClojureCL, ClojureCUDA and Neanderthal. He shows how a dynamic REPL workflow can be combined with low-level hardware acceleration for machine learning and data science.

Why HN recommends it

HN commenters point to it as the talk by the author of Neanderthal and the Clojure GPU libraries, including comparisons against the JVM and CPU.

Watch his talk from EuroClojure 2016 in Bratislava — HN, 2017

Functional programming · Machine learning & deep learning · Computer architecture & CPUs

Subtitles (excerpt)

okay so uh my name is dragon jurich and welcome to closure is not afraid of the GPU and um just let me start with uh a few details about me I am a professor of sare engineering at University of Belgrade and surprisingly I actually love doing uh programming so I've been doing this uh I've been using closure as my primary programming language since 2009 and I also managed to put it into the curriculum at University so I teach it since 2010 and uh you can find my contacts uh on GitHub and the the projects that I will uh be talking about today uh could be uh the documentation could be find found at uncomplicated today uh is on the slide so please relax and enjoy the talk uh you don't have to read every word on the slides uh so let's uh start a bit uh so I've seen that Rich hii has a hammock uh and I also have a hammock this is my hammock now uh I guess in Stephen King's uh guide about Java these four would be some four main vain or so now you remember the first time uh that you saw lisp many years ago uh I suppose and there were those lisp guys talking about their language lisp and how it's awesome and when you looked at it it was like oh it's weird but now when you uh accommodated yourself to the

On Hacker News

ClojureCUDA, ClojureCL (as in OpenCL) and Neanderthal are libraries developed by Dragan Djuric. Watch his talk from EuroClojure 2016 in Bratislava: [VIDEO] — skardan, 2017-05
ClojureCL looks quite awesome, it looks like it takes all of the boiler plate out in comparison to C but I am very interested in the overhead. I tried aparapi for my bachelor thesis and for larger problems it seemed to have quite the overhead that the normal kernel did not have (several years ago). After watching your talk on youtube [1], I saw you compared the GPU implementation to the JVM and o… — thebeardedone, 2018-02
Data crunching on GPUs is a LISP strength. It runs code orders of magnitude faster than say python or c++. [VIDEO] — tmpLisp, 2016-12

Related videos