Keynote Speech

 

 

Facets of Publish/Subscribe for Large-scale Information Dissemination Applications

(And How To Effectively Filter Information For Millions Of Subscribers)

 

Hans-Arno Jacobsen

Department of Electrical and Computer Engineering & Department of Computer Science

University of Toronto

jacobsen@eecg.toronto.edu

http://www.eecg.toronto.edu/~jacobsen

 

 

 

Abstract

Publish/subscribe is a messaging paradigm that permits information producers to selectively disseminate information and permits information consumers to filter out irrelevant information.  This paradigm has wide applicability ranging from selective information dissemination and location-based services to workload management. 

The challenges in modeling and building distributed systems with this paradigm lie in designing appropriate publish/subscribe middleware systems that can support large numbers of subscribers (e.g., in the millions), support highly volatile subscriber interests, support high rates of information input (e.g., thousands of events per minute), and support varying content formats (e.g., text-based, table-based, and semi-structured.) 

In this talk, I will survey the research activities towards addressing these challenges undertaken by the Middleware System Research Group at the University of Toronto in the context of the Toronto Publish/Subscribe System Project (ToPSS).  I will discuss emerging application scenarios, analyze their requirements, sketch the design of our ToPSS middleware system research prototype, present algorithmic results, and outline our research agenda in this context.  This will draw from the A-ToPSS (approximate matching-based ToPSS), the L-ToPSS (location-based ToPSS), the S-ToPSS (semantic matching-based ToPSS), and the mobility-aware distributed publish/subscribe system sub-projects. 

More information about ToPSS can be found at: http://www.eecg.toronto.edu/~jacobsen

 

Short Bio:

Hans-Arno Jacobsen holds a faculty position with the Department of Electrical and Computer Engineering and the Department of Computer Science at the University of Toronto.  His principal areas of research include middleware systems, distributed systems, and information systems.  Dr. Jacobsen received his Ph.D. degree with highest honors (summa cum laude) from the Institute of Information Systems, Humboldt University, Berlin in 1999.  His research was sponsored by a doctoral fellowship awarded through the German Research Foundation for participation in the graduate program on "Distributed Information Systems" in Berlin.  His research was honored with the annual IBM Best Paper Award and his dissertation was awarded the Humboldt Prize. 

Prior to his doctoral research, Hans-Arno Jacobsen worked for over three years in research laboratories in the US, including the International Computer Science Institute (ICSI) in Berkeley and the Lawrence Berkeley National Laboratory (LBNL).  He received his M.S. Degree in Computer Science from the University of Karlsruhe, Germany in 1995.  During his undergraduate studies, he participated in various exchange student programs, visiting the Computer Science Department at the University of California, Berkeley, the Institute of Applied Mathematics in Grenoble, France, and the Department of Psychology at the University Louis Pasteur in Strasbourg, France. 

Hans-Arno Jacobsen has served as program committee member for the International Conference on Distributed Computing Systems, the International Symposium on Distributed Objects and Applications, the Very Large Database Conference, and the Middleware Conference, and has been involved in the organization of many recent international workshops. 

More information is available from http://www.eecg.toronto.edu/~jacobsen