• DocumentCode
    3317128
  • Title

    A dynamic load balancing algorithm for sort-first rendering clusters

  • Author

    Hui, Chang ; Xiaoyong, Lei ; Shuling, Dai

  • Author_Institution
    Sch. of Autom. Sci. & Electr. Eng., Beihang Univ., Beijing, China
  • fYear
    2009
  • fDate
    8-11 Aug. 2009
  • Firstpage
    515
  • Lastpage
    519
  • Abstract
    Load balancing determines the performance and scalability of sort-first rendering clusters. A dynamic load balancing algorithm based on deferred shading is presented. The proposed algorithm is based on the 2-pass rendering character of deferred shading to predict the rendering load distribution among the screen space. Utilizing fragment shaderings and alpha blending in GPU, the proposed algorithm can achieve per-pixel load distribution as a gray image in real-time. The paper also presented the screen space division strategy and the optimization methods to the proposed algorithm. Experiments show that the proposed algorithm can achieve dynamic load-balancing with little additional computation. Compared with other load balancing strategies, the proposed algorithm can distribute rendering load more uniformly among the rendering node and improve system´s render performance prominently.
  • Keywords
    pattern clustering; rendering (computer graphics); resource allocation; alpha blending; dynamic load balancing algorithm; fragment shadering; graphical processing unit; optimization method; rendering load distribution; screen space division strategy; sort-first rendering cluster; Clustering algorithms; Heuristic algorithms; Load management; Load modeling; Optimization methods; Partitioning algorithms; Rendering (computer graphics); Resource management; Scalability; Space technology; deferred shading; load-balancing; parallel rendering; rendering clusters;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Information Technology, 2009. ICCSIT 2009. 2nd IEEE International Conference on
  • Conference_Location
    Beijing
  • Print_ISBN
    978-1-4244-4519-6
  • Electronic_ISBN
    978-1-4244-4520-2
  • Type

    conf

  • DOI
    10.1109/ICCSIT.2009.5234831
  • Filename
    5234831