UTS: An Unbalanced Tree Search Benchmark

  • Stephen Olivier
  • Jun Huan
  • Jinze Liu
  • Jan Prins
  • James Dinan
  • P. Sadayappan
  • Chau-Wen Tseng
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4382)

Abstract

This paper presents an unbalanced tree search (UTS) benchmark designed to evaluate the performance and ease of programming for parallel applications requiring dynamic load balancing. We describe algorithms for building a variety of unbalanced search trees to simulate different forms of load imbalance. We created versions of UTS in two parallel languages, OpenMP and Unified Parallel C (UPC), using work stealing as the mechanism for reducing load imbalance. We benchmarked the performance of UTS on various parallel architectures, including shared-memory systems and PC clusters. We found it simple to implement UTS in both UPC and OpenMP, due to UPC’s shared-memory abstractions. Results show that both UPC and OpenMP can support efficient dynamic load balancing on shared-memory architectures. However, UPC cannot alleviate the underlying communication costs of distributed-memory systems. Since dynamic load balancing requires intensive communication, performance portability remains difficult for applications such as UTS and performance degrades on PC clusters. By varying key work stealing parameters, we expose important tradeoffs between the granularity of load balance, the degree of parallelism, and communication costs.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Harris, T.: The Theory of Branching Processes. Springer, Heidelberg (1963)MATHGoogle Scholar
  2. 2.
    Eastlake, D., Jones, P.: US secure hash algorithm 1 (SHA-1). Internet Engineering Task Force, RFC 3174 (Sept. 2001), Available at http://www.rfc-editor.org/rfc/rfc3174.txt
  3. 3.
    Leskovec, J., Kleinberg, J., Faloutsos, C.: Graphs over time: densification laws, shrinking diameters and possible explanations. In: Proc. 11th ACM SIGKDD Int’l Conf. Know. Disc. Data Mining (KDD ’05), pp. 177–187. ACM Press, New York (2005)CrossRefGoogle Scholar
  4. 4.
    Blumofe, R., Leiserson, C.: Scheduling multithreaded computations by work stealing. In: Proc. 35th Ann. Symp. Found. Comp. Sci., Nov., pp. 356–368 (1994)Google Scholar
  5. 5.
    Kumar, V., Grama, A.Y., Vempaty, N.R.: Scalable load balancing techniques for parallel computers. J. Par. Dist. Comp. 22(1), 60–79 (1994)CrossRefGoogle Scholar
  6. 6.
    Kumar, V., Rao, V.N.: Parallel depth first search. part II. analysis. Int’l J. Par. Prog. 16(6), 501–519 (1987)MATHCrossRefMathSciNetGoogle Scholar
  7. 7.
    UPC Consortium. UPC language specifications, v1.2. Lawrence Berkeley National Lab, Tech. Rep. LBNL-59208 (2005)Google Scholar
  8. 8.
    Frigo, M., Leiserson, C.E., Randall, K.H.: The implementation of the Cilk-5 multithreaded language. In: Proc. 1998 SIGPLAN Conf. Prog. Lang. Design Impl (PLDI ’98), pp. 212–223 (1998)Google Scholar
  9. 9.
    Marowka, A.: Analytic comparison of two advanced c language-based parallel programming models. In: Proc. Third Int’l Symp. Par. and Dist. Comp./Int’l Workshop Algorithms, Models and Tools for Par. Comp. Hetero. Nets (ISPDC/HeteroPar’04), pp. 284–291 (2004)Google Scholar
  10. 10.
    Marathe, J., Mueller, F.: Hardware profile-guided automatic page placement for ccnuma systems. In: Proc. 11th ACM SIGPLAN Symp. Princ. Pract. Par. Prog (PPoPP ’06), pp. 90–99. ACM Press, New York (2006)CrossRefGoogle Scholar
  11. 11.
    Berlin, K., et al.: Evaluating the impact of programming language features on the performance of parallel applications on cluster architectures. In: Rauchwerger, L. (ed.) LCPC 2003. LNCS, vol. 2958, pp. 194–208. Springer, Heidelberg (2004)Google Scholar
  12. 12.
    European Center for Parallelism. PARAVER (2006), Available at http://www.cepba.upc.edu/paraver/
  13. 13.
    Chen, W., Iancu, C., Yelick, K.A.: Communication optimizations for fine-grained UPC applications. In: Proc. Int’l Conf. Par. Arch. Compilation Tech. (PACT 2005) pp. 267–278 (2005)Google Scholar
  14. 14.
    Prins, J., et al.: UPC implementation of an unbalanced tree search benchmark. Univ. North Carolina at Chapel Hill, Tech. Rep. TR03-034 (Oct. 2003)Google Scholar

Copyright information

© Springer Berlin Heidelberg 2007

Authors and Affiliations

  • Stephen Olivier
    • 1
  • Jun Huan
    • 1
  • Jinze Liu
    • 1
  • Jan Prins
    • 1
  • James Dinan
    • 2
  • P. Sadayappan
    • 2
  • Chau-Wen Tseng
    • 3
  1. 1.Dept. of Computer Science, Univ. of North Carolina at Chapel Hill 
  2. 2.Dept. of Computer Science and Engineering, The Ohio State Univ. 
  3. 3.Dept. of Computer Science, Univ. of Maryland at College Park 

Personalised recommendations