Skip to content Skip to main navigation Report an accessibility issue

Graph Computing at Scale

SukumarDr. Sreenivas Rangan Sukumar, Research Scientist
Computational Science and Engineering Division, ONRL
January  16th, 2015, 2:30 – 3:30 PM
410 John D. Tickle Building 

Sreenivas Rangan Sukumar is a data scientist and a machine learning researcher developing knowledge nurturing architectures and algorithms. As a data enthusiast with access to several unique computing infrastructures at leadership computing resources at the Oak Ridge National Lab, he designs and implements data-driven knowledge discovery algorithms (in-database, in-memory and in-situ) that can handle massive-scale datasets (order of petabytes). He has over 40 publications (book chapters, journals, technical reports, peer-reviewed conference and workshop proceedings) in areas of disparate data collection, organization, processing, integration, fusion, analysis and inference. His work has been applied to a wide variety of domains such as healthcare, social network analysis, electric grid modernization and public policy informatics. He received his Bachelor’s degree in Electronics and Communication Engineering from the University of Madras, India in 2002, a Master’s and a Doctor of Philosophy degree in Electrical Engineering from the University of Tennessee, Knoxville in 2004 and 2008 respectively. His graduate work included topics in photo-realistic 3D scene imaging, 3D/2D computer vision, image processing, mobile robotics and his current research interests are in data science, science of data, pattern recognition, graph computing, data fusion for machine intelligence.

Talk Abstract: The URiKA appliance is a new addition to scalable graph computing at ORNL that provides researchers the opportunity for staging, querying and semantically reasoning with massive heterogeneous graphs. In the past year, we have been developing, implementing and testing graph algorithms at scale and observing several orders of magnitude speed-up compared to graph analysis using relational and graph databases. In this seminar, we introduce the tools, capabilities, APIs and algorithms that have been developed applied to knowledge discovery in different scientific domains. We will be presenting use-cases in pattern search, healthcare, semantic reasoning with knowledgebases, literature-based discovery and the insights derived using unconventional but now possible graph-data analysis at scale.