Distributed skyline query processing
Event details
| Date | 14.05.2009 |
| Hour | 11:00 |
| Speaker | Dr. Akrivi Vlachou, Norwegian University of Science and Technology (NTNU) |
| Location | |
| Category | Conferences - Seminars |
Abstract: Skyline queries have attracted much attention recently, mainly because they help users to make intelligent decisions over data that represent many conflicting criteria. A challenging problem is to support skyline queries in distributed environments, where data is scattered over independent sources. First, we investigate skyline query processing over peer-to-peer (P2P) systems, which emerge as a powerful model for searching vast amounts of data distributed over independent sources. Similar to the peer-to-peer file-sharing networks (eMule, KaZaA), our approach relies on a super-peer architecture. In order to support query processing, queries are first routed on the super-peer backbone and afterwards, if necessary, they are forwarded to the peers with relevant data. Query processing in P2P networks poses inherent challenges and demands non-traditional techniques due to the distribution of content and the lack of global knowledge. Our approach relies on routing information at super-peer level in a deliberate way, aiming at minimizing contacted peers and transferred data. Then, we study efficient skyline processing assuming a different distributed environment, namely a parallel shared-nothing architecture. The main difference from peer-to-peer systems is that we do not aim to minimize the number of contacted peers, but minimize the total response time by distributing evenly the workload among all participating servers. Finally, we discuss the case of a mobile environment, where the bandwidth consumption should be minimized.
Short Bio: Akrivi Vlachou is currently a post-doctoral researcher under an ERCIM “Alain Bensoussan” fellowship in the database research group of Prof.Kjetil Nørvåg, in NTNU, Norway. She recieved her PhD in 2008 entitled "Efficient Query Processing over Highly Distributed Data" from the Athens University of Economics and Business supervised by Prof.Michalis Vazirgiannis. She received her Master in "Advanced Information Systems" and her Diploma in “Computer Science and Telecommunications” from the Department of Computer Science, University of Athens in 2001 and 2003 respectively. Her research interests lie in query processing and data management in large-scale distributed systems.
Practical information
- General public
- Free