Elsevier

Performance Evaluation

Volume 33, Issue 3, August 1998, Pages 201-213
Performance Evaluation

On analytic modeling of multimedia batching schemes

https://doi.org/10.1016/S0166-5316(98)00017-0Get rights and content

Abstract

In modern multimedia-on-demand systems, user requests for video programs can be serviced in batches. An important aspect in this type of service is providing access to the users upon request, either instantaneously or within a small latency. We develop an approximate model for evaluating the performance of multimedia batching schemes with respect to system latency. Our mathematical model provides a method for analyzing the performance of the first come first served (FCFS) and the round robin (RR) scheme. We also study the maximal queue length (MQL) scheme and derive a bound on the ratio of its expected latency to optimal. Finally, we give a simple algorithm called the rounded ratio. We show that its expected latency is at most twice optimal.

References (22)

  • W. Sincoskie

    System architecture for large scale video on demand

    Comput. Networks ISDN Systems

    (1991)
  • S. Aggarwal et al.

    Adaptive video on demand

  • C.C. Aggarwal et al.

    On optimal batching policies for video-on-demand storage servers

  • S. Aggarwal et al.

    On optimal piggyback merging policies for VOD systems

  • D. Anderson

    Metascheduling for continuous media

    ACM Trans. Comput. Sys.

    (1993)
  • A. Dan et al.

    Scheduling policies for on-demand video server with batching

  • H. Dykeman et al.

    Scheduling algorithms for videotex systems under broadcast deliver

  • Electronic Engineering Times

    (1993)
  • E. Fox

    The coming revolution in interactive digital video

    Commun. ACM

    (1989)
  • E. Gelenbe

    G-networks with triggered customer movement

    J. Appl. Prob.

    (1993)
  • E. Gelenbe

    G-networks with signals and batch removals

    Prob. Engrg. Inform. Sci.

    (1993)
  • Cited by (0)

    Part of this work was done while the author was with IBM T.J. Watson Research Center, Yorktown Heights, NY 10598, USA.

    View full text