Event composition and detection in data stream management systems

  • Authors:

  • Mukesh Mohania;Dhruv Swamini;Shyam Kumar Gupta;Sourav Bhowmick;Tharam Dillon

  • Affiliations:

  • IBM India Research Lab, I.I.T., Hauz Khas, New Delhi;Dept of Computer Science and Engg, I.I.T. Delhi, Hauz Khas, New Delhi;Dept of Computer Science and Engg, I.I.T. Delhi, Hauz Khas, New Delhi;Nanyang Technological University, Singapore;Faculty of Information Technology, University of Technology Sydney, Australia

  • Venue:
  • DEXA'05 Proceedings of the 16th international conference on Database and Expert Systems Applications
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

There has been a rising need to handle and process streaming kind of data. It is continuous, unpredictable, time-varying in nature and could arrive in multiple rapid streams. Sensor data, web clickstreams, etc. are the examples of streaming data. One of the important issues about streaming data management systems is that it needs to be processed in real-time. That is, active rules can be defined over data streams for making the system reactive. These rules are triggered based on the events detected on the data stream, or events detected while summarizing the data or combination of both. In this paper, we study the challenges involved in monitoring events in a Data Stream Management System (DSMS) and how they differ from the same in active databases. We propose an architecture for event composition and detection in a DSMS, and then discuss an algorithm for detecting composite events defined on both the summarized data streams and the streaming data.