Baradwaj Wins 2006 Nancy G. Pollock Thesis Award
Congratulations to Nikhil Baradwaj (MS ’05), who has been selected by the NC State University Graduate School as the recipient of the 2006 Nancy G. Pollock Thesis Award.
Sponsored by the NC State University Graduate School, the Nancy G. Pollock Thesis and Dissertation Award program is designed to reward outstanding scholarly research and to demonstrate the positive impact of graduate-level research on both the economy of North Carolina and the quality of life for all its citizens. The MS Thesis Award carries a $500 stipend.
Fellow alum, Lina Battestilli, won the 2006 Nancy G. Pollock Dissertation Award. Both are the first graduate students from computer science ever to receive these prestigious awards.
Baradwaj conducted his MS research under the guidance and supervision of Dr. George Rouskas, and his work was funded by a grant from the National Science Foundation.
His thesis entitled, "Traffic Quantization and its Application to QoS Routing", investigated traffic quantization as a method for supporting per-flow functionality in packet-switched networks in an efficient and scalable manner. Traffic quantization naturally leads to the concept of tiered service which has the potential to alleviate the complexity associated with supporting per-flow quality of service (QoS). Unlike a conventional continuous-rate network, a quantized network offers only a small set of service tiers (levels). The main motivation for offering a tiered service is to simplify a wide range of network functions, including traffic policing, packet scheduling, traffic engineering, state dissemination, network management, service level agreements, billing, etc.
As part of his research, Baradwaj introduced and defined the directional p-median problem as a general framework for modeling and reasoning about a wide range of quantization problems in several computer science domains. He characterized the complexity of the directional p-median problem in one and multiple dimensions, and presented efficient optimal algorithms and/or robust heuristics to solve it.
He demonstrated the benefits of quantization by presenting efficient and simple solutions for certain important algorithms whose implementation is quite complex and involved in non-quantized contexts. He also supported the overall findings with a combination of theoretical and experimental results.
Born in New Delhi, India, Baradwaj completed his MS in Computer Science at NC State University in August 2005. He is now employed with MicroStrategy, in McLean, Virginia.
Return To News Homepage