Monday, February 25, 2008

Cloud Computing: Project Overviews

Jimmy Lin
University of Maryland

11am, February 27, 2008
A.V. Williams 2120

Abstract

The University of Maryland is one of six U.S. universities involved in a new initiative lead by Google/IBM to explore the MapReduce programming paradigm, through an open-source implementation called Hadoop. Using MapReduce, programmers can harness enormous amounts of computing power centralized in large data centers. The initiative aims to provide faculty and students with access to next-generation "cloud computing" technologies, allowing them to think at "Web scale".

The Spring 2008 "cloud computing" course represents an attempt to integrate teaching and research in the context of this Google/IBM initiative. The basic idea is to bring together graduate and undergraduate students to tackle large-data problems, focused on the areas of natural language processing and information retrieval. This semester, twelve students are working on five different projects:

  • Statistical Machine Translation
  • Language Modeling
  • Identity Resolution in Email Archives
  • Retrieval in the Biomedical Domain
  • Text-Background Separation in Picture Books
  • Biological Sequence Alignment

After a very brief introduction, students in the cloud computing course will provide an overview of their projects and outline what they plan to accomplish this semester. Additional information can be found at the projects overview page.

Contributors