• DocumentCode
    3040470
  • Title

    Quantitative Analysis of Partition Statistics and their Impact on Data Replication in MANETs

  • Author

    Shi, Ke ; Haas, Zygmunt J.

  • Author_Institution
    Huazhong Univ. of Sci. & Tech., Hubei
  • fYear
    2007
  • fDate
    5-8 July 2007
  • Firstpage
    48
  • Lastpage
    48
  • Abstract
    Data replication is, indeed, a widely used redundancy scheme for improving data accessibility in distributed systems at the cost of increased storage space and communication overhead. In this paper, we study the performance of data replication under mobile ad hoc networking environments in the presence of frequent network partitions by analytical modeling and by statistical analysis of simulation results. In particular, we examine the statistics of network partitions for a number of mobility models, and we propose distribution models to approximate the size of the network partitions. We then establish the relation between the network partitioning pattern and the effectiveness of the data replication scheme, which could be used to dynamically adjust the degree of replication depending on the current network operational conditions, while optimizing the trade-off between storage and data accessibility.
  • Keywords
    ad hoc networks; mobile radio; statistical analysis; MANET; data accessibility; data replication; distributed systems; distribution models; mobile ad hoc networking environments; mobility models; network partitioning pattern; partition statistics; quantitative analysis; statistical analysis; Analytical models; Availability; Costs; Equations; Mobile ad hoc networks; Mobile communication; Performance analysis; Redundancy; Statistical analysis; Statistical distributions;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Parallel and Distributed Computing, 2007. ISPDC '07. Sixth International Symposium on
  • Conference_Location
    Hagenberg
  • Print_ISBN
    0-7695-2917
  • Type

    conf

  • DOI
    10.1109/ISPDC.2007.43
  • Filename
    4271938