A Proposal for Task-Generating Loops in OpenMP*

  • Xavier Teruel
  • Michael Klemm
  • Kelvin Li
  • Xavier Martorell
  • Stephen L. Olivier
  • Christian Terboven
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8122)


With the addition of the OpenMP* tasking model, programmers are able to improve and extend the parallelization opportunities of their codes. Programmers can also distribute the creation of tasks using a worksharing construct, which allows the generation of work to be parallelized. However, while it is possible to create tasks inside worksharing constructs, it is not possible to distribute work when not all threads reach the same worksharing construct. We propose a new worksharing-like construct that removes this restriction: the taskloop construct. With this new construct, we can distribute work when executing in the context of an explicit task, a single, or a master construct, enabling us to explore new parallelization opportunities in our applications. Although we focus our current work on evaluating expressiveness rather than performance evaluation, we present some initial performance results using a naive implementation for the new taskloop construct based on a lazy task instantiation mechanism.


OpenMP Task Worksharing Loop Fork/Join 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Ayguadé, E., Copty, N., Duran, A., Hoeflinger, J., Lin, Y., Massaioli, F., Teruel, X., Unnikrishnan, P., Zhang, G.: The Design of OpenMP Tasks. IEEE Trans. Parallel Distrib. Syst. 20(3), 404–418 (2009)CrossRefGoogle Scholar
  2. 2.
    Balart, J., Duran, A., Gonzàlez, M., Martorell, X., Ayguadé, E., Labarta, J.: Nanos Mercurium: a Research Compiler for OpenMP. In: Proc. of the 6th European Workshop on OpenMP (EWOMP 2004), pp. 103–109 (October 2004)Google Scholar
  3. 3.
    Blumofe, R.D., Leiserson, C.E.: Scheduling Multithreaded Computations by Work Stealing. Journal of the ACM 46(5), 720–748 (1999)MathSciNetzbMATHCrossRefGoogle Scholar
  4. 4.
    Ferrer, R.: Task Chunking of Iterative Constructions in OpenMP 3.0. In: Proc. of the 1st Workshop on Execution Environments for Distributed Computing, pp. 49–54 (July 2007)Google Scholar
  5. 5.
    Ferrer, R., Duran, A., Martorell, X., Ayguadé, E.: Unrolling Loops Containing Task Parallelism. In: Gao, G.R., Pollock, L.L., Cavazos, J., Li, X. (eds.) LCPC 2009. LNCS, vol. 5898, pp. 416–423. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  6. 6.
    Kurzak, J., Ltaief, H., Dongarra, J.J., Badia, R.M.: Scheduling for Numerical Linear Algebra Library at Scale. In: Proc. of the High Performance Computing Workshop, pp. 3–26 (June 2008)Google Scholar
  7. 7.
    Leiserson, C.E.: The Cilk++ Concurrency Platform. The Journal of Supercomputing 51(3), 244–257 (2010)CrossRefGoogle Scholar
  8. 8.
    Microsoft: Task Parallel Library (2013), (last accessed June 21, 2013)
  9. 9.
    OpenMP Architecture Review Board: OpenMP Application Program Interface, Version 3.1 (July 2011)Google Scholar
  10. 10.
    OpenMP Architecture Review Board: OpenMP Application Program Interface, Version 4.0: Public Review Release Candidate 2 (March 2013)Google Scholar
  11. 11.
    Reinders, J.: Intel Threading Building Blocks. O’Reilly, Sebastopol (2007)Google Scholar
  12. 12.
    Terboven, C., Schmidl, D., Cramer, T., an Mey, D.: Task-Parallel Programming on NUMA Architectures. In: Kaklamanis, C., Papatheodorou, T., Spirakis, P.G. (eds.) Euro-Par 2012. LNCS, vol. 7484, pp. 638–649. Springer, Heidelberg (2012)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Xavier Teruel
    • 1
  • Michael Klemm
    • 2
  • Kelvin Li
    • 3
  • Xavier Martorell
    • 1
  • Stephen L. Olivier
    • 4
  • Christian Terboven
    • 5
  1. 1.Barcelona Supercomputing CenterSpain
  2. 2.Intel CorporationUSA
  3. 3.IBM CorporationUSA
  4. 4.Sandia National LaboratoriesUSA
  5. 5.RWTH Aachen UniversityGermany

Personalised recommendations