The Big Data buzzword represents the convergence of massive, dynamic data sets with powerful techniques for effective use of that data. At the core is a scale-out architecture involving distributed computation and storage.
Machine learning algorithms are adapting to work effectively in distributed environments. Eron will examine a few frameworks and tools for scale-out machine learning algorithms. We will discuss how the Hadoop framework is evolving to support a greater diversity of machine learning algorithms. We will touch on the use of the cloud for hosting scale-out machine learning projects.
The session will consist of a presentation, some demonstrations, and open discussion.
About the speaker:
Eron Wright is a Director of Engineering at EMC. Eron works on the ViPR product, delivering data services for petabyte-scale storage with integrated Hadoop capabilities. Eron was formerly a developer at Microsoft, working on the Windows Azure platform. In his spare time, he is a machine learning hobbyist, and welcomes our new sentient overlords!