Subscribe to our Erlang Factory newsletter to receive the latest updates and news

Prashanth Mundkur
Disco Hacker @ Nokia

Prashanth Mundkur is a distributed systems researcher at Nokia. He is a core developer and maintainer of the Disco computing framework.

Prashanth Mundkur is Giving the Following Talks
Beyond MapReduce

The goal of Disco has been to be a simple and usable implementation of MapReduce. To keep things simple, this MapReduce aspect has been hard-coded into Disco, both in the Erlang job scheduler, as well as in the Python library. To fix various issues in the implementation, we decided to take a cold hard look at the dataflow in Disco's version of MapReduce. We came up with a generalization that should be more flexible and hence also more useful than plain old MapReduce. We call this the Pipeline model, and we hope to use this in the next major release of Disco. This will implement the old MapReduce model in terms of a more general programmable pipeline, and also expose the pipeline to users wishing to take advantage of the optimization opportunities it offers.

If time permits, we will also discuss other aspects of the Disco roadmap, and the future of the Disco project.

Talk objectives: Introduce features in the upcoming release of Disco, and talk about their design and motivation, and how Disco users could take advantage of them.

Target audience: Big data hackers and Disco users.