• DocumentCode
    3230003
  • Title

    OMFS: an object-oriented multimedia file system for cluster streaming server

  • Author

    Cheng, Bin ; Jin, Hai ; Yuan, Quan

  • Author_Institution
    Cluster & Grid Comput. Lab., Huazhong Univ. of Sci. & Technol., Wuhan
  • fYear
    2005
  • fDate
    1-1 July 2005
  • Lastpage
    537
  • Abstract
    Using cluster as streaming servers is now a very popular solution. With the improvement of network speed and CPU performance, file I/O becomes the key bottleneck of cluster streaming server. However, most existing systems are short of an efficient and uniform multimedia file system so that the designers of cluster streaming server have to take a lot of time in managing all kinds of multimedia files. In this paper we propose an object-oriented multimedia file system, called OMFS, to provide an object-oriented organization scheme for multimedia files and uniform interfaces for cluster streaming servers. Furthermore, we present some optimization strategies to improve its performance, such as bypassing kernel buffer, creating time-based block indexing map. Our experimental results show that OMFS not only reduces the start-up latency and VCR waiting time but also improves the throughput of cluster streaming server. In addition, our results demonstrate that object-oriented organization framework is efficient and feasible to simplify the design of cluster streaming server
  • Keywords
    file organisation; media streaming; multimedia servers; object-oriented programming; cluster streaming server; object-oriented multimedia file system; Delay; File servers; File systems; Indexing; Kernel; Multimedia systems; Network servers; Streaming media; Throughput; Video recording;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    High-Performance Computing in Asia-Pacific Region, 2005. Proceedings. Eighth International Conference on
  • Conference_Location
    Beijing
  • Print_ISBN
    0-7695-2486-9
  • Type

    conf

  • DOI
    10.1109/HPCASIA.2005.57
  • Filename
    1592317