3rd International ICST Conference on Quality of Service in Heterogeneous Wired/Wireless Networks

Research Article

Multiuser prefetching with queuing prioritization in heterogeneous wireless systems

  • @INPROCEEDINGS{10.1145/1185373.1185417,
        author={Ben Liang and Stephen Drew},
        title={Multiuser prefetching with queuing prioritization in heterogeneous wireless systems},
        proceedings={3rd International ICST Conference on Quality of Service in Heterogeneous Wired/Wireless Networks},
        publisher={ACM},
        proceedings_a={QSHINE},
        year={2006},
        month={8},
        keywords={mobile prefetching heterogeneous wireless networks WLAN/3G integration performance modelling queuing analysis},
        doi={10.1145/1185373.1185417}
    }
    
  • Ben Liang
    Stephen Drew
    Year: 2006
    Multiuser prefetching with queuing prioritization in heterogeneous wireless systems
    QSHINE
    ACM
    DOI: 10.1145/1185373.1185417
Ben Liang1,*, Stephen Drew1,*
  • 1: Department of Electrical and Computer Engineering, University of Toronto, 10 King’s College Road, Toronto, Ontario, M5S 3G4, Canada
*Contact email: liang@comm.utoronto.ca, drews@comm.utoronto.ca

Abstract

We study the performance of a multi-user prefetching strategy in a two-tier heterogeneous wireless network. A predictive framework was previously introduced for mobility-aware document prefetching to enhance the experience of a mobile user roaming between heterogeneous wireless access networks. However, an undesirable effect of multiple prefetching users is the potential for system instability due to the racing behavior between document access delay and user prefetch quantity. This phenomenon is particularly acute in the heterogeneous environment. We propose to alleviate the system traffic load through optimizing a prefetch thresholding algorithm, accounting for server queuing prioritization. We evaluate the performance of the proposed algorithm through numerical analysis and simulation. We show that stability can be maintained even under heavy usage, providing both the same scalability as a non-prefetching system and the performance gains associated with prefetching.