State aware data dissemination over structured overlays

Liping Chen, Gul Agha

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We describe the problem of data dissemination in stream-oriented applications where the required filter is a function of the current state. We call such functions dynamic filters. A State Aware Data Dissemination Network (SA-DDN) is proposed to support dynamic filters. Two approaches Single-level Filtering (SF) and Multilevel Filter Decomposition (MFD) are proposed to facilitate the data dissemination. We show how MFD improves performance over SF. We then describe a realization of SA-DDN on top of an improved bi-directional Chord overlay with a built-in multicast mechanism. An application of stock price monitoring is implemented based on SA-DDN and real life stock quotes are collected to demonstrate the feasibility of our system. Extensive simulations are performed to compare the performance of both approaches and provide insight into the advantages of MFD.

Original languageEnglish (US)
Title of host publicationProceedings - 6th IEEE International Conference on Peer-to-Peer Computing, P2P 2006
EditorsAlberto Montresor, Adam Wierzbicki, Nahid Shahmehri
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages145-152
Number of pages8
ISBN (Electronic)0769526799, 9780769526799
DOIs
StatePublished - 2006
Event6th IEEE International Conference on Peer-to-Peer Computing, P2P 2006 - Cambridge, United Kingdom
Duration: Sep 6 2006Sep 8 2006

Publication series

NameProceedings - 6th IEEE International Conference on Peer-to-Peer Computing, P2P 2006

Conference

Conference6th IEEE International Conference on Peer-to-Peer Computing, P2P 2006
Country/TerritoryUnited Kingdom
CityCambridge
Period9/6/069/8/06

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Hardware and Architecture
  • Safety, Risk, Reliability and Quality

Fingerprint

Dive into the research topics of 'State aware data dissemination over structured overlays'. Together they form a unique fingerprint.

Cite this