CSR-DMSS, SM: Design and Evaluation of a Scalable Meta-Event Dissemination System

Iowa State University, Ames IA

Investigators

Abstract

CSR-DMSS, SM: Design and Evaluation of a Scalable Meta-Event Dissemination System Project Abstract: Event dissemination systems offer a powerful middleware for distributed applications such as electronic trading platforms, news aggregators, and network intrusion detection systems. Such systems are usually based on the publish-subscribe paradigm: published events are delivered to clients who have a supplied a matching subscription. While most current publish-subscribe systems support subscriptions that evaluate to true/false on a per-event basis, emerging applications need support for subscriptions to higher level "meta-events", which correspond to combinations of basic events. The challenge in tracking such subscriptions is that there is no single event, when processed in isolation, permits the evaluation of a subscription to true or false. This project designs, implements and evaluates an event dissemination system that supports tracking of meta-events. The system converts user-defined subscriptions into "triggers", and efficiently maintains the state of the triggers using algorithms inspired by the literature on data stream processing. The system scales with increasing numbers of subscriptions through merging triggers by identifying commonalities between them, and through indexing triggers for fast detection of matching subscriptions for events. The system is evaluated using an internet-scale testbed, such as Planetlab. Such a meta-event dissemination system will provide a more expressive interface to the designer of distributed and web applications, thus transforming the way in which distributed applications are authored. This is expected to impact the application of information technology in various fields such as healthcare, finance, and security. The project enhances undergraduate and graduate education through strengthening the freshmen engineering course in Electrical and Computer Engineering and through developing a strong algorithmic foundation for research students in networking and distributed computing.