Managing Extreme-Scale Information Processing via Performance Modeling, Analysis and Distributed Methods

Speaker:
Prof. Cathy H. Xia
Department of Integrated Systems Engineering, The Ohio State University

Time:
2:00pm, 2014-10-17

Bio:
Dr. Cathy H. Xia is an associate professor in the Department of Integrated Systems Engineering at The Ohio State University, with an adjunct appointment from Computer Science and Engineering. Her area of expertise is stochastic modeling, analysis and distributed optimization with applications in computer networks and information systems. Dr. Xia received her Ph.D. in Operations Research from Stanford University and has worked at IBM T.J. Watson Research Center as a research scientist for many years. She is the co-editor of book “Performance Modeling and Engineering”, and has published over 60 peer-reviewed journal articles and conference proceedings.

Abstract:
With emerging applications such as cloud computing, big data analytics, smart grids, etc., modern information and service computing networks are growing increasingly complex and large scale. In this talk, we present new performance models and analysis, and distributed methods to study the scalable design and efficient management of future extreme-scale information networks. We will first present a fork-join queueing network model and study the scalability of large-scale information processing systems. We then present new first-order and second-order approaches for deriving distributed control algorithms in managing large-scale information networks with guaranteed fast convergence.