On Dynamic Load Balancing on Graphics Processors (English)

In: Graphics Hardware   ;  57-64  ;  2008

How to get this document?

Download
Commercial Copyright fee: €14.50 Basic fee: €4.00 Total price: €18.50
Academic Copyright fee: €4.50 Basic fee: €2.00 Total price: €6.50

To get maximum performance on the many-core graphics processors it is important to have an even balance of the workload so that all processing units contribute equally to the task at hand. This can be hard to achieve when the cost of a task is not known beforehand and when new sub-tasks are created dynamically during execution. With the recent advent of scatter operations and atomic hardware primitives it is now possible to bring some of the more elaborate dynamic load balancing schemes from the conventional SMP systems domain to the graphics processor domain. We have compared four different dynamic load balancing methods to see which one is most suited to the highly parallel world of graphics processors. Three of these methods were lock-free and one was lock-based. We evaluated them on the task of creating an octree partitioning of a set of particles. The experiments showed that synchronization can be very expensive and that new methods that take more advantage of the graphics processors features and capabilities might be required. They also showed that lock-free methods achieves better performance than blocking and that they can be made to scale with increased numbers of processing units.

  • Title:
    On Dynamic Load Balancing on Graphics Processors
  • Author / Creator:
  • Published in:
  • Publisher:
    The Eurographics Association
  • Place of publication:
    Postfach 8043, 38621 Goslar, Germany
  • Year of publication:
    2008
  • Size:
    8 pages
  • ISBN:
  • ISSN:
  • DOI:
  • Type of media:
    Conference paper
  • Type of material:
    Electronic Resource
  • Language:
    English
  • Source:
  • Export:
  • ORKG:

Table of contents conference proceedings

The tables of contents are generated automatically and are based on the data records of the individual contributions available in the index of the TIB portal. The display of the Tables of Contents may therefore be incomplete.

1
Tracy: A Debugger and System Analyzer for Cross-Platform Graphics Development
Kyöstilä, Sami / Kangas, Kari J. / Pulli, Kari | 2008
13
Total Recall: A Debugging Framework for GPUs
Sharif, Ahmad / Lee, Hsien-Hsin S. | 2008
21
A Hardware Processing Unit for Point Sets
Heinzle, Simon / Guennebaud, Gaël / Botsch, Mario / Gross, Markus | 2008
33
Coherent Layer Peeling for Transparent High-Depth-Complexity Scenes
Carr, Nathan / Mech, Radomir / Miller, Gavin | 2008
41
Non-Uniform Fractional Tessellation
Munkberg, Jacob / Hasselgren, Jon / Akenine-Möller, Tomas | 2008
47
All-Pairs Shortest-Paths for Large Graphs on the GPU
Katz, Gary J. / Jr., Joseph T. Kider | 2008
57
On Dynamic Load Balancing on Graphics Processors
Cederman, Daniel / Tsigas, Philippas | 2008
65
GPU Accelerated Pathfinding
Bleiweiss, Avi | 2008
75
Floating-Point Buffer Compression in a Unified Codec Architecture
Ström, Jacob / Wennersten, Per / Rasmusson, Jim / Hasselgren, Jon / Munkberg, Jacob / Clarberg, Petrik / Akenine-Möller, Tomas | 2008
85
DHTC: An Effective DXTC-based HDR Texture Compression Scheme
Sun, Wen / Lu, Yan / Wu, Feng / Li, Shipeng | 2008
95
An Improved Shading Cache for Modern GPUs
Sitthi-amorn, Pitchaya / Lawrence, Jason / Yang, Lei / Sander, Pedro V. / Nehab, Diego | 2008