Filter Similarities in Content-Based Publish/Subscribe Systems

  • Authors:
  • Gero Mühl;Ludger Fiege;Alejandro P. Buchmann

  • Affiliations:
  • -;-;-

  • Venue:
  • ARCS '02 Proceedings of the International Conference on Architecture of Computing Systems: Trends in Network and Pervasive Computing
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

Matching notifications to subscriptions and routing notifications from producers to interested consumers are the main problems in large-scale publish/subscribe systems.Most previously proposed distributed notification services either use flooding or, if filtering is performed, they assume that each event broker has global knowledge about all active subscriptions. Both approaches degrade the scalability of notification services as the former wastes network resources and the latter generates overly large routing tables.In this paper we describe content-based routing algorithms that exploit filter similarities in order to reduce the size of routing tables and the number of control messages that are exchanged among the brokers in order to keep the routing tables up-to-date. In particular, the proposed algorithms do not assume global knowledge about all active subscriptions. Furthermore, we describe how these optimizations can be supported if the underlying data and filter model is based on structured records.