• DocumentCode
    341824
  • Title

    VBR video data scheduling using window-based prefetching

  • Author

    Kim, In-Hwan ; Kim, Jeong-Won ; Lee, Seung-Won ; Chung, Ki-Dong

  • Author_Institution
    Dept. of Comput. Sci., Pusan Nat. Univ., South Korea
  • Volume
    1
  • fYear
    1999
  • fDate
    36342
  • Firstpage
    159
  • Abstract
    The MPEG data used in a video server has variable bit rate (VBR) characteristics because each frame is different in size, and the difference of movements between frames is also variable. So, a video server must consider this VBR characteristic in resource management. So far, for VBR data, many scheduling schemes have been proposed. When VBR video is displayed, if disk bandwidth is allocated assuming the worst case, the utilization of disk bandwidth becomes very low. In order to raise the utilization of disk bandwidth, a prefetching scheme is used, but this needs extra buffers. We propose a WVS (Window-based VBR data scheduling scheme) for VBR data. WVS prefetches data with a limited maximum bandwidth for each window unit. Moreover, WVS, which considers the state of system resources, can obtain a tradeoff between disk bandwidth and buffer allocation. The proposed scheme constructs in advance CTL (Constant Time Length) meta data about a video and dynamically allocates resources for each window unit. In simulations, we compared the proposed scheme with previous schemes in terms of the size of resource allocation and the number of concurrent users
  • Keywords
    bandwidth allocation; meta data; multimedia communication; resource allocation; scheduling; variable rate codes; video coding; video on demand; video servers; MPEG; VBR video data scheduling; buffer allocation; buffers; concurrent users; constant time length meta data; disk bandwidth allocation; limited maximum bandwidth; resource allocation; resource management; simulation; variable bit rate; video frame; video server; window-based prefetching; Bandwidth; Bit rate; Buffer storage; Computer science; Information retrieval; Prefetching; Processor scheduling; Resource management; Transform coding; Video compression;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Multimedia Computing and Systems, 1999. IEEE International Conference on
  • Conference_Location
    Florence
  • Print_ISBN
    0-7695-0253-9
  • Type

    conf

  • DOI
    10.1109/MMCS.1999.779140
  • Filename
    779140