Enhanced DR-Tree for Low Latency Filtering in Publish/Subscribe Systems

  • Authors:
  • Luciana Arantes;Maria Gradinariu Potop-Butucaru;Pierre Sens;Mathieu Valero

  • Affiliations:
  • -;-;-;-

  • Venue:
  • AINA '10 Proceedings of the 2010 24th IEEE International Conference on Advanced Information Networking and Applications
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Distributed R-tree overlays emerged as an alternative for efficiently implementing DHT-free publish/subscribe communication primitives. Overlays using R-tree index structures offer logarithmic delivery garantis, guarantee zero false negatives and considerably reduce the number of false positives. In this paper we extend the distributed R-trees (DR-trees) in order to reduce event delivery latency. Our optimizations target both the structural organization of the DR-Trees and the publication policies. The contribution of the current work steams in an extensive evaluation of the novel structure along four parameters: latency, load, scalability and the rate of false positives. The enhanced structure performs better than the traditional distributed R-tree in terms of delivery latency. Additionally, it does not alter the performances related to the scalability, nor the load balancing of the tree, and neither the rate of false positives and negatives filtered by a node.