MapReduce with CloudHaskell: Difference between revisions
Julianporter (talk | contribs) (New page: Category:ApplicationsCategory:MonadCategory:LibrariesCategory:ConcurrencyCategory:ParallelCategory:Research ==Introduction== This is documentation of my work on d...) |
Julianporter (talk | contribs) |
||
| Line 16: | Line 16: | ||
Code to be placed on Hackage when more robust. | Code to be placed on Hackage when more robust. | ||
=== | ===Distributed job scheduler=== | ||
Not yet started. | Not yet started. | ||
[[User:Julianporter|julianporter]] 18:20, 31 October 2011 (UTC) | [[User:Julianporter|julianporter]] 18:20, 31 October 2011 (UTC) | ||
Revision as of 23:00, 31 October 2011
Introduction
This is documentation of my work on developing a proof-of-concept demonstrator for MapReduce using CloudHaskell to provide a framewok for distributed applications, and the monadic approach to MapReduce described here.
Status
Storage
I have developed a very simple distributed storage service that provides what is needed to ship data to each processing node at the start of each round of processing, and then to assemble their outputs ready to form the input for the next round.
- Description of what I have done here
- A working application can be found at git://github.com/Julianporter/Distributed-Haskell.git
Code to be placed on Hackage when more robust.
Distributed job scheduler
Not yet started.
julianporter 18:20, 31 October 2011 (UTC)