Abstract: in this talk I introduce the concept of Quality-of-Information-Aware networking. Most communication network theories, designs and control algorithms address performance metrics such as throughput, delay or errors, in terms of data bits. We postulate that communication networks should be viewed as information sources, and thus should be evaluated and controlled in terms of the quality of information they convey. We consider information metrics such as completeness, accuracy, precision and timeliness. Many of these metrics must be specified in the context in which the information is being used. We consider both end-device processing and network transfer when processing QoI-aware information requests. In this talk I define QoI, provide an overview of the research ongoing in the Army Research Lab-funded Network Science research program, show specific results from the research, and discuss current problems.
Bio: Thomas F. La Porta is the William E. Leonhard Chair Professor in the Computer Science
and Engineering Department at Penn State. He received his B.S.E.E. and M.S.E.E. degrees from
The Cooper Union, New York, NY, and his Ph.D. degree in Electrical Engineering from
Columbia University, New York, NY. He joined Penn State in 2002. He is the Director of the
Institute of Networking and Security Research at Penn State. Prior to joining Penn State, Dr. La
Porta was with Bell Laboratories since 1986. There, he was the Director of the Mobile
Networking Research Department where he led various projects in wireless and mobile
networking. He is an IEEE Fellow, Bell Labs Fellow, received the Bell Labs Distinguished
Technical Staff Award, and an Eta Kappa Nu Outstanding Young Electrical Engineer Award. He
also won a Thomas Alva Edison Patent Awards in 2005 and 2009. Dr. La Porta was the founding
Editor-in-Chief of the IEEE Transactions on Mobile Computing. He also served as Editor-in-
Chief of IEEE Personal Communications Magazine.
Abstract: In designing distributed computing systems, conundrums abound:
* Should we aim to balance load among servers, or is that a convenient myth?
* Is it better to use 1 fast server, or many slow ones?
* Does a closed-loop system behave the same as a open-loop one if run with the same load?
* Does high load imply high delay? Does low load imply low delay?
* Does favoring short jobs starve the long ones?
* How useful is job replication in high-variability cloud settings?
* Should one turn off servers when not in use to save power?
* If the arrival rate increases, should one scale up the number of servers by the same factor increase?
This talk will answer these and other questions using fun, humor, and lots of prizes!
Bio: Mor Harchol-Balter is a professor of computer science at Carnegie Mellon University. From 2008-2011, she served as the Associate Department Head. Mor received her doctorate from the Computer Science department at the University of California at Berkeley under the direction of Manuel Blum in 1996 and then spent three years at MIT under the NSF Postdoctoral Fellowship in the Mathematical Sciences. She is a recipient of the McCandless Chair, the NSF CAREER award, multiple best paper awards, and several teaching awards, including the Herbert A. Simon Award for Teaching Excellence. Mor is heavily involved in the ACM SIGMETRICS research community, where she served as Technical Program Chair for Sigmetrics 2007 and as General Chair for Sigmetrics 2013. Mor's work focuses on designing new resource allocation policies (load balancing policies, power management policies, and scheduling policies) for server farms and distributed systems in general. She is known for both her work in queueing analysis and in systems implementation. She has co-authored over 100 publications in top journals and conferences, including a textbook, Performance Analysis and Design of Computer Systems , published by Cambridge University Press, 2013, and has given several keynote talks in top conferences. Mor has graduated many PhD students, the majority of whom are now professors in top academic institutions.