is the easiest way to share online.
Learn more »
Bioinformatics in the Near Concurrent Future -
March 5, 2009
5 other people
wooo, I need to look at Clojure now.... -
a Lisp dialect, you say?
[Of topic] Lol. Saw "Manneken pis around the world" on that blog post. There's only one manneken pis, though... [/of topic] -
Clojure is definitely very popular in some sectors (shows up on Hacker News all the time) -
looking at Clojure, I think I'll wait until PyPy is done and supports concurrency
If you want pure and fast concurrency, Haskell seems to be the best bet. Clojure seems to be a great way to access all the existing Java technology while writing in something that feels much more contemporary. -
An addendum. In order to have a snow ball's chance in hell of managing genomic datasets, algorithims are going to need to run asymmetric multi-core environments like GPUs and Cell. I've heard rumors that Google is working on asymmetric multi-core JVM. But otherwise young bioinfomaticians should be learning NVDIA/AMD's C frameworks or Haskell. -
Zaki, I am not so sure. The accelerator space is still playing out. As long as you keep the programming models as they are, that will remain a specialized place. Plus, for most bioinformatics applications, you need scale out architectures, not scale up (anything that's not numerical heavy essentially) -
©2013 FriendFeed -
Tools & Widgets