Presenter: Ioannis Kamitsos
Date: 20 July 2016

This talk will focus on a few recent research results on greening of the Information and Communication Technology. In the first part, I will demonstrate how resource pooling can be exploited to optimize the tradeoff between two intrinsically conflicting objectives in cloud computing – energy consumption and delay performance. I will then present how we extend our optimal energy conserving techniques to the case of DSL broadband access and show how our approach results in a more energy efficient and stable DSL operation compared to existing Broadband Forum power saving policies. Finally, I will demonstrate an analytic framework that helps data center operators minimize day-to-day operational costs (such as electricity and bandwidth costs), based on processing jobs characteristics, as well as optimally expand their data center network based on predictions of future operational costs.

Ioannis Kamitsos is a Senior Software Engineer with Bloomberg LP, New York. He received the Diploma degree in electrical and computer engineering from the National Technical University of Athens, Greece, in 2006, and the M.A. and Ph.D. degrees from Princeton University, Princeton, NJ, USA, in 2009 and 2012, respectively. During the summer of 2010, he interned at the Standards and RF Laboratory, Samsung Telecommunications America, Richardson, TX, USA. His research interests include optimization theory, optimal control, and machine learning/big data analytics.