Algorithm design for resource allocation in stream computing and multi-radio multi-channel wireless networks

---

Algorithm design for resource allocation in stream computing and multi-radio multi-channel wireless networks

Speaker:

Dr. Srinivasan Parthasarathy
Exploratory Stream Analytics group
IBM Research

Time: 3PM-4PM Wednesday, October 13, 2010
Location: Babbio 304, Stevens Institute of Technology, Hoboken, NJ 07030

Abstract:

Stream computing is an important paradigm within the realm of high-performance computing, where the goal is continuous analysis of massively high-volume data-in-motion in order to support ultra low-latency decision-making. In the first part of my talk, I will deal with the problem of optimizing the evaluation of conjunctive queries on streaming data. This problem arises in a variety of streaming applications such as multimedia classification, semantic publish-subscribe systems, and network traffic monitoring. I will describe algorithm design techniques for minimizing the cost of conjunctive query evaluation in a sequential (single-processor) environment. I will also describe a generic method for transforming such cost (or latency) minimization algorithms developed in the context of single-processor environments, into algorithms for flow (or throughput) maximization in parallel processing environments. In the second part of the talk, I will present distributed and local-control algorithms for joint channel allocation and link scheduling in multi-channel multi-radio wireless networks. A central theme of this talk is the design and analysis of algorithms with provably good performance guarantees. Results from this work have appeared in various conferences including ACM SIGMOD (2008), ACM PODS (2008), and IEEE INFOCOM (2009).

Bio Sketch

Srinivasan Parthasarathy is a Research Staff Member in the Exploratory Stream Analytics group at IBM Research. His research interests are centered around algorithm design and optimization on the one hand, and a variety of application areas including stream computing, wireless / ad hoc networking, Internet computing, information retrieval, and query optimization on the other. Prior to joining IBM Research, he received his Ph.D. (2006) and M.S. (2003) degrees in computer science from the University of Maryland at College Park, and B. Tech. degree (2000) in computer science and engineering from the Indian Institute of Technology (IIT Madras). His research publications span several top journals and conferences in Computer Science such as the Journal of the ACM (2009 & 2006), IEEE/ACM Transactions on Networking (2008), IEEE FOCS (2002 & 2005), ACM Sigmetrics (2005), and ACM-SIAM SODA (2004).

---

---

---

---