Dynamic MapReduce on IBM Platform Computing

Jian Tan, research staff member at IBM T.J. Watson Research Center, will present “Dynamic MapReduce on IBM Platform Computing,” on Tuesday, Feb. 11 at 11 a.m. in 3129 Shelby Center. The seminar, which focuses on IBM’s programming tool Dynamic MapReduce, or DynMR, is hosted by Auburn University’s Department of Computer Science and Software Engineering.

DynMR aims to resolve various performance challenges that persist in MapReduce, a programming tool for processing large data sets. Tan will discuss resolutions made by IBM Platform Computing to the following MapReduce operations: difficulty in selecting optimal performance parameters for a single job in a fixed, dedicated environment, as well as complete inability to configure parameters that can perform optimally in a dynamic, multi-job cluster; long job execution resulting from a task long-tail effect, often caused by ReduceTask data skew or heterogeneous computing nodes; inefficient use of hardware resources as a result of ReduceTask bundling several functional phases together which may idle during certain phases of execution.

Jian Tan
Research Staff Member, IBM

Tan is a research staff member at IBM T. J. Watson Research Center in Yorktown Heights, NY. He received master's and doctoral degrees in electrical engineering from Columbia University in 2004 and 2008, respectively, as well as a bachelor's degree from the University of Science and Technology of China, in 2002. His doctoral thesis work won the Eliahu Jury Award from Columbia University, and he was a postdoctoral researcher with the Networking and Communications Research Lab at The Ohio State University from 2009 to 2010. He interned with Lucent Bell Laboratories during in 2005 and 2006, and with Microsoft Research at Cambridge, UK, in 2007. His current research interests focus on systems and theoretical modeling for large-scale distributed computing.

Tuesday, February 11, 2014, 11:00 am - 12:00 pm
3129 Shelby Center