• DocumentCode
    660657
  • Title

    Resolve Hotspots and Load Imbalance Problem in Event Stream Processing System

  • Author

    BaoJian Zhou ; Zhongzhi Luan ; Jieqian Wu ; Ming Xie

  • Author_Institution
    Sino-German Joint Software Inst., Beihang Univ., Beijing, China
  • fYear
    2013
  • fDate
    4-6 Nov. 2013
  • Firstpage
    158
  • Lastpage
    159
  • Abstract
    Event stream processing is the system which uses the processing element to process the data in streams of an application. The processing element (PE) is the computing unit that can handle and compute the metadata in the stream queue. However, there are two problems. First problem is that hotspots maybe happen when applications run in cluster environment; the second problem is load imbalance in the cluster when hotspots appear in that system. Therefore, in this paper, we mainly research how the hotspots happen and how to resolve the imbalance problem caused by hotspots.
  • Keywords
    distributed processing; meta data; resource allocation; cluster environment; event stream processing system; hotspots; load imbalance problem; metadata; processing element; stream queue; Benchmark testing; Cloud computing; Clustering algorithms; Computational modeling; Load modeling; Prototypes; Radiation detectors; Event Stream; Hotspot; Load Imbalance; PE;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Cloud and Service Computing (CSC), 2013 International Conference on
  • Conference_Location
    Beijing
  • Type

    conf

  • DOI
    10.1109/CSC.2013.34
  • Filename
    6693196