Colloquium Details

User-Centric Data Processing in Resource Constraint Environments

Speaker: Mohamed Sharaf, University of Toronto

Location: Warren Weaver Hall 102

Date: April 30, 2009, 11:30 a.m.

Host: Michael Overton


In our increasingly pervasive environment, users, mobile or stationary, have performance expectations which define the success of any application. Mostly, such applications are data-intensive which makes the timely processing of queries very challenging, particularly given the various resource constraints of the different environments such as limitations in power and wireless communication as in mobile and sensor applications, or CPU and memory as in data stream processing systems.

In this talk, I will highlight few of the resource allocation challenges in the three main stages of data management in pervasive environments, namely: (1) data acquisition, (2) data processing, and (3) data dissemination. Further, I will present several solutions which address the efficient allocation of those resources. In particular, I will discuss the TiNA scheme for energy-aware data processing in sensor networks. I will also present my work on query scheduling in both data stream management systems and web database systems within the context of the AQSIOS project. Moreover, I will talk about the TraQIOS scheme for optimizing I/O-intensive applications in shared data centers. Finally, I will describe my recent work on location-aware data broadcasting services in mobile and wireless networks.

Speaker Bio:

Dr. Sharaf is currently a Postdoctoral Research Fellow at the University of Toronto. He received his Ph.D. in Computer Science from the University of Pittsburgh in 2007. He was the recipient of the Taulbee Award for Excellence in Computer Science in 2002 and a two-time winner of the Andrew Mellon Predoctoral Fellowship in 2003 and 2004. In 2008, he was awarded a two-year post-doctoral fellowship from the Ontario Ministry of Research and Innovation (MRI).



Refreshments will be offered starting 15 minutes prior to the scheduled start of the talk.

How to Subscribe