Event processing in the form of Event-Condition-Action rules has been researched extensively from the situation monitoring viewpoint to detect changes in a timely manner and to take appropriate actions. Several event specification languages and processing models have been developed, analyzed, and implemented. Lately, data stream processing has been receiving a lot of attention to deal with applications that generate large amounts of data in real-time at varying input rates and to compute functions over multiple streams that satisfy quality of service (QoS) requirements. A few systems based on the data stream processing model have been proposed to deal with change detection and situation monitoring. However, current data stream processing models lack the notion of composite event specification and computation, and they cannot be readily combined with event detection and rule specification, which are necessary and important for many applications.

The goal of this research project is to facilitate a synergistic integration of stream processing with event processing to meet the needs of new class of applications (e.g., intelligent home monitoring, accident monitoring on highways). The concept of window in stream processing is generalized and enhanced to a semantic window that can be applied to both streams and events. Quality of service (QoS) requirements is addressed in the context of event processing (using scheduling and load shedding strategies) to obtain a seamless end-to-end system. Current time-based semantics of event operators is extended to include attribute-based semantics to increase the expressiveness of event operators. Optimization of event operators in the presence of attribute-based semantics is addressed as well. A proof-of-principle system to incorporate the results of this research is implemented. Extensive performance and scalability experiments are conducted using data from linear road benchmark and intelligent home monitoring (e.g., MavHome) to evaluate the effectiveness of the algorithms and the implementation of the system. The results of this research will have far reaching effects on sensor as well as pervasive computing applications such as environmental monitoring, RFID- and GIS-based tracking and monitoring systems.

By synthesizing these two models and combining their strengths, we argue that the integrated model will be better than the sum of its parts. In this research, we analyze the similarities and differences between the event and data stream processing models and investigate the following tasks:

• Generalize the notion of a semantic window, which significantly extends the current window concept for continuous queries (CQs), and stream modifiers in order to extend current stream computation model for complicated change detection. Investigate and develop load shedding strategies.
• Extend event specification language for handling stream data. Extend event processing model for handling data streams with high input rates by generalizing consumption modes.
• Develop and implement an integrated model that combines the capabilities of both models for applications that not only need to monitor changes through CQs, but also to express and process complex events generated by CQs.

Year 1 tasks:

• Theoretical analysis of load shedding and its implications on accuracy and quality of service (QoS) requirements – specifically tuple latency,

• Location of load shedders in the overall stream processing architecture to reduce overhead,

• algorithm for optimal placement of load shedders to minimize erro using random as well as semantic approaches, and

• implementation of runtime optimizer that includes load shedders.

Year 2 Tasks:

•  Semantic windows,

• Efficient computation of semantic windows,  and

• Generalization of event operators to include attribute-based semantics

Year 3 Tasks:

• Issues related to scheduling and buffer management in event processing including the relationship between consumption modes and windows.

• Completion of MavEStream implementation

• Experimental analysis (for performance and scalability)