Advertisement

A Combined Technique of Non-uniform Loops

  • Sam Jin Jeong
  • Kun Hee Han
  • Young Chul Park
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3947)

Abstract

This paper proposes efficient methods such as Improved Tiling Method for non-uniform dependence loops with only flow dependences and Improved Region Partitioning Method for loops with both flow and anti dependences. In the Improved Tiling Method, we propose our incrementing minimum dependence distance technique and loop interchanging technique. In Improved Region Partitioning Method, we eliminate anti dependences from the nested loop by variable renaming. After variable renaming, this method can divide the iteration space into two parallel regions as large as possible and one or less serial region as small as possible. By combination of existing methods and our proposed methods, it exposes more parallelism.

Keywords

Extreme Point Iteration Space Nest Loop Dependence Tail Combine Technique 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Lilja, D.J.: Exploiting the parallelism available in loop. IEEE Computer, 13–26 (1994)Google Scholar
  2. 2.
    Tzen, T., Ni, L.: Dependence uniformization: A loop parallelization technique. IEEE Trans. Parallel and Distributed Systems 4(5), 547–558 (1993)CrossRefGoogle Scholar
  3. 3.
    Punyamurtula, S., Chaudhary, V.: Minimum dependence distance tiling of nested loops with non-uniform dependences. In: Proc. Symp. Parallel and Distributed Processing, pp. 74–81 (1994)Google Scholar
  4. 4.
    Ju, J., Chaudhary, V.: Unique sets oriented Partitioning of nested loops with non-uniform dependences. In: Proc. Int. Conf. Parallel Processing, vol. III, pp. 45–52 (1996)Google Scholar
  5. 5.
    Zaafrani, A., Ito, M.R.: Parallel region execution of loops with irregular dependencies. In: Proc. Int. Conf. Parallel Processing, vol. II, pp. 11–19 (1994)Google Scholar
  6. 6.
    Cho, C.K., Lee, M.H.: A Loop Parallization Method for Nested Loops with Non-uniform Dependences. In: Proceedings of the International Conference on Parallel and Distributed Systems, pp. 314–321 (1997)Google Scholar
  7. 7.
    Jeong, S.J.: Maximizing Parallelism for Nested Loops with Non-uniform Dependences. In: Laganá, A., Gavrilova, M.L., Kumar, V., Mun, Y., Tan, C.J.K., Gervasi, O. (eds.) ICCSA 2004. LNCS, vol. 3046, pp. 213–222. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  8. 8.
    Pean, D.-L., Chen, C.: CDCHP: a new effective mechanism to maximize parallelism of nested loops with non-uniform dependences. The Journal of Systems and Software 56, 279–297 (2001)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Sam Jin Jeong
    • 1
  • Kun Hee Han
    • 1
  • Young Chul Park
    • 1
  1. 1.Division of Information and Communication EngineeringCheonan UniversityCheonan CityKorea

Personalised recommendations