MavEStream: Synergistic Integration of Stream and Event Processing

Abstract

Although research seems to address event and stream data processing as two separate topics, there are a number of similarities between them. For many advanced stream applications, both event and rule processing are needed and are not currently well-supported. Extant event processing systems concentrate primarily on complex events and rules and stream processing systems concentrate on stream operators, scheduling, and quality of service issues. Synergistic integration of these models will be better than the sum of its parts. We propose an integrated model to combine the capabilities of both models for applications that need both of them. Specifically, we introduced a number of enhancements, including stream modifiers, semantic windows, event generators, and enhanced event and rule specifications, to couple two models seamlessly. We prototype our integrated system using the stream processing system (MavStream) with the event processing system (Snoop and Sentinel) and discuss the design and implementation issues of our prototype.