Net-χ: unified data-centric internet services

  • Authors:
  • Praveen Rao;Justin Cappos;Varun Khare;Bongki Moon;Beichuan Zhang

  • Affiliations:
  • Department of Computer Science, University of Arizona, Tucson, AZ;Department of Computer Science, University of Arizona, Tucson, AZ;Department of Computer Science, University of Arizona, Tucson, AZ;Department of Computer Science, University of Arizona, Tucson, AZ;Department of Computer Science, University of Arizona, Tucson, AZ

  • Venue:
  • NETB'07 Proceedings of the 3rd USENIX international workshop on Networking meets databases
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Databases and networks currently have different service models. Database services are data-centric in that users typically describe the content of data and the system finds and returns matching data. However, traditional Internet services are server-centric in that users have to know the location of data (e.g., a URL) in order to retrieve it. We envision a future in which Internet services are data-centric. Users specify their interests and publishers describe their data. Based on the matching between user interests and data contents, users can pull data from publishers, and publishers can push data to interested users. We propose a unified system design called Net-χ to support data-centric Internet services seamlessly under a common framework. In Net-χ, user interests and data contents are characterized by polynomial signatures. These signatures are stored in a distributed hash table, on which interest matching is performed. Users can download matching data from publishers and publishers can push data to many interested users via per-document data-driven dissemination trees. By leveraging a wide range of database and networking techniques, Net-χ provides a scalable, flexible, and secure infrastructure for data-centric Internet services.