Definition of a simple MAP Reduce API

Question

I am developing a distributed processing API in Java, Erlang and Ruby. What basic commands can I include from which I can build mapreduce, pipelining, and all the most used parallell algorithms on top of it.

Answer 1

You may get some hints from this blog post, which I found quite useful.

http://ox.no/posts/minimalistic-mapreduce-in-net-4-0-with-the-new-task-parallel-library-tpl