CISS: an efficient object clustering framework for DHT-Based peer-to-peer applications

  • Authors:
  • Jinwon Lee;Hyonik Lee;Seungwoo Kang;Sungwon Choe;Junehwa Song

  • Affiliations:
  • Division of Computer Science, Department of Electrical Engineering & Computer Science, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea;Division of Computer Science, Department of Electrical Engineering & Computer Science, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea;Division of Computer Science, Department of Electrical Engineering & Computer Science, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea;Division of Computer Science, Department of Electrical Engineering & Computer Science, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea;Division of Computer Science, Department of Electrical Engineering & Computer Science, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea

  • Venue:
  • DBISP2P'04 Proceedings of the Second international conference on Databases, Information Systems, and Peer-to-Peer Computing
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Distributed Hash Tables (DHTs) have been widely adopted in many Internet-scale P2P systems. Emerging P2P applications such as massively multi player online games (MMOGs) and P2P catalog systems frequently update data or issue multi-dimensional range queries, but existing DHT-based P2P systems can not support these applications efficiently due to object declustering. Object declustering can result in significant inefficiencies in data update and multi-dimensional range query routing. In this paper, we propose CISS, a framework that supports efficient object clustering for DHT-based P2P applications. While utilizing DHT as a basic lookup layer, CISS uses a Locality Preserving Function (LPF) instead of a hash function. Thus, CISS achieves a high level of clustering without requiring any changes to existing DHT implementations. Technically, we study LPF encoding function, efficient routing protocols for data updates and multi-dimensional range queries, and cluster-preserving load balancing. We demonstrate the performance benefits of CISS through simulation.