A predictive approach to achieving consistency in cooperative caching in MANET

Yu  Huang; Jiannong  Cao; Beihong  Jin

1st International ICST/Create-Net Workshop on Peer-to-Peer Information Management

Research Article

A predictive approach to achieving consistency in cooperative caching in MANET

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1145/1146847.1146898,
    author={Yu  Huang and Jiannong  Cao and Beihong  Jin},
    title={A predictive approach to achieving consistency in cooperative caching in MANET},
    proceedings={1st International ICST/Create-Net Workshop on Peer-to-Peer Information Management},
    publisher={ACM},
    proceedings_a={P2PIM},
    year={2006},
    month={6},
    keywords={},
    doi={10.1145/1146847.1146898}
}

Yu Huang
Jiannong Cao
Beihong Jin
Year: 2006
A predictive approach to achieving consistency in cooperative caching in MANET
P2PIM
ACM
DOI: 10.1145/1146847.1146898

Yu Huang^1,2^,*, Jiannong Cao²^,*, Beihong Jin³^,*

1: Dept. of Computer Science, Univ. of Science and Technology of China, Hefei, China.
2: Internet and Mobile Computing Lab, Dept. of Computing, Hong Kong Polytechnic Univ., Kowloon, Hong Kong.
3: Technology Center of Software Engineering, Institute of Software, Chinese Academy of Sciences, Beijing, China.

*Contact email: csyhuang@comp.polyu.edu.hk, csjcao@comp.polyu.edu.hk, jbh@otcaix.iscas.ac.cn

Abstract

Cooperative caching is a very important technique for efficient data dissemination and sharing in mobile ad hoc networks (MANETs). Many applications have requirements on the consistency of the content cached on different nodes. However, this issue has not been adequately addressed and few of the existing solutions are really adaptive in a dynamic MANET environment. In this paper, we propose a predictive algorithm called PCC (Predictive Caching Consistency) that can make on-line tradeoff between the level of consistency of cached data and the overhead associated with achieving it. Simulation results show that PCC works adaptively and efficiently in a MANET environment, even with dynamic changes in the size of the network, mobility model of nodes, number of queries, and frequency of data updates.

Published: 2006-06-01
Publisher: ACM

: http://dx.doi.org/10.1145/1146847.1146898