Performance analysis of cache consistency strategies for multi-hop wireless networks

  • Authors:
  • Wenzhong Li;Edward Chan;Daoxu Chen;Sanglu Lu

  • Affiliations:
  • State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China 210093;Department of Computer Science, City University of Hong Kong, Kowloon, Hong Kong;State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China 210093;State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China 210093

  • Venue:
  • The Journal of Supercomputing
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Data caching is widely used to facilitate information access in wireless mobile networks. Caching frequently used content on mobile clients brings benefits by reducing bandwidth usage, saving energy, and leveraging server workload. Maintaining cache consistency is an important issue in mobile caching and has received much attention by researchers. However, most existing studies on cache consistency strategies are limited to cellular wireless networks in which the mobile terminals can communicate with the base stations directly. In multi-hop wireless network environment, mobile terminals communicate with data server through multiple unreliable links with a high probability of disconnection from the network, which makes it more challenging for the maintenance of cache data consistency. In this paper, we investigate three cache consistency strategies for multi-hop wireless networks: the pull-based strategy POD (pull on demand), the push-based strategy MAT (modified amnesic terminals) and the store and forward strategy PIR (pull-based invalidation report). Extensive theoretical analysis is provided to compare the cache hit ratio, query delay and communication overhead of these methods, and the impact of system parameters on performance is studied as well. Simulation experiments are used to examine the performance of these three schemes, and it is shown that PIR provides the best overall performance.