Work Distribution of Data-Parallel Applications on Heterogeneous Systems

  • Suejb MemetiEmail author
  • Sabri Pllana
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9945)


Heterogeneous computing systems offer high peak performance and energy efficiency, and utilizing this potential is essential to achieve extreme-scale performance. However, optimal sharing of the work among processing elements in heterogeneous systems is not straightforward. In this paper, we propose an approach that uses combinatorial optimization to search for optimal system configuration in a given parameter space. The optimization goal is to determine the number of threads, thread affinities, and workload partitioning, such that the overall execution time is minimized. For combinatorial optimization we use the Simulated Annealing. We evaluate our approach with a DNA sequence analysis application on a heterogeneous platform that comprises two Intel Xeon E5 processors and an Intel Xeon Phi 7120P co-processor. The obtained results demonstrate that using the near-optimal system configuration, determined by our algorithm based on the simulated annealing, application performance is improved.


Execution Time Simulated Annealing Heterogeneous System System Configuration Simulated Annealing Algorithm 
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.


  1. 1.
    TOP500 Supercomputer Sites. Accessed Jan 2016
  2. 2.
    Abraham, E., Bekas, C., Brandic, I., Genaim, S., Johnsen, E.B.,Kondov, I., Pllana, S., Streit, A.: Preparing HPC applications for exascale: challenges and recommendations. In: 2015 18th International Conference on Network-Based Information Systems (NBiS), pp. 401–406, September 2015Google Scholar
  3. 3.
    Albayrak, O.E., Akturk, I., Ozturk, O.: Improving application behavior on heterogeneous manycore systems through kernel mapping. Parallel Comput. 39(12), 867–878 (2013). CrossRefGoogle Scholar
  4. 4.
    Augonnet, C., Thibault, S., Namyst, R., Wacrenier, P.A.: StarPU: a unified platform for task scheduling on heterogeneous multicore architectures. Concurr. Comput. Pract. Exp. 23(2), 187–198 (2011)CrossRefGoogle Scholar
  5. 5.
    Ayguadé, E., Blainey, B., Duran, A., Labarta, J., Martínez, F., Martorell, X., Silvera, R.: Is the Schedule clause really necessary in OpenMP? In: Voss, M.J. (ed.) WOMPAT 2003. LNCS, vol. 2716, pp. 147–159. Springer, Heidelberg (2003). doi: 10.1007/3-540-45009-2_12 CrossRefGoogle Scholar
  6. 6.
    Benkner, S., Pllana, S., Traff, J., Tsigas, P., Dolinsky, U., Augonnet, C., Bachmayer, B., Kessler, C., Moloney, D., Osipov, V.: PEPPHER: efficient and productive usage of hybrid computing systems. IEEE Micro 31(5), 28–41 (2011)CrossRefGoogle Scholar
  7. 7.
    Braun, T.D., Siegel, H.J., Beck, N., Bölöni, L.L., Maheswaran, M., Reuther, A.I., Robertson, J.P., Theys, M.D., Yao, B., Hensgen, D., et al.: A comparison of eleven static heuristics for mapping a class of independent tasks onto heterogeneous distributed computing systems. J. Parallel Distrib. Comput. 61(6), 810–837 (2001)CrossRefzbMATHGoogle Scholar
  8. 8.
    Chrysos, G.: Intel\(\textregistered \) Xeon Phi Coprocessor-the Architecture. Intel Whitepaper (2014)Google Scholar
  9. 9.
    Dokulil, J., Bajrovic, E., Benkner, S., Pllana, S., Sandrieser, M.,Bachmayer, B.: High-level support for hybrid parallel execution of C++ applications targeting Intel Xeon Phi coprocessors. In: ICCS. Procedia Computer Science, vol. 18, pp. 2508–2511. Elsevier (2013)Google Scholar
  10. 10.
    Duran, A., Ayguadé, E., Badia, R.M., Labarta, J., Martinell, L., Martorell, X., Planas, J.: OmpSs: a proposal for programming heterogeneous multi-core architectures. Parallel Process. Lett. 21(02), 173–193 (2011)MathSciNetCrossRefGoogle Scholar
  11. 11.
    Grewe, D., O’Boyle, M.F.P.: A static task partitioning approach for heterogeneous systems using OpenCL. In: Knoop, J. (ed.) CC 2011. LNCS, vol. 6601, pp. 286–305. Springer, Heidelberg (2011). doi: 10.1007/978-3-642-19861-8_16 CrossRefGoogle Scholar
  12. 12.
    Kessler, C.W., Dastgeer, U., Thibault, S., Namyst, R., Richards, A., Dolinsky, U., Benkner, S., Traff, J.L., Pllana, S.: Programmability and performance portability aspects of heterogeneous multi-/manycore systems, pp. 1403–1408. IEEE (2012)Google Scholar
  13. 13.
    Khan, F., Han, Y., Pllana, S., Brezany, P.: Estimation of parameters sensitivity for scientific workflows. In: 2009 International Conference on Parallel Processing Workshops, pp. 457–462, September 2009Google Scholar
  14. 14.
    Khan, F., Han, Y., Pllana, S., Brezany, P.: An ant-colony-optimization based approach for determination of parameter significance of scientific workflows. In: 2010 24th IEEEInternational Conference on Advanced Information Networking and Applications (AINA), pp. 1241–1248, April 2010Google Scholar
  15. 15.
    Kołodziej, J., Khan, S.U.: Data scheduling in data grids and data centers: a short taxonomy of problems and intelligent resolution techniques. In: Nguyen, N.-T., Kołodziej, J., Burczyński, T., Biba, M. (eds.) Transactions on Computational Collective Intelligence X. LNCS, vol. 7776, pp. 103–119. Springer, Heidelberg (2013). doi: 10.1007/978-3-642-38496-7_7 CrossRefGoogle Scholar
  16. 16.
    Liu, Y., Pan, T., Aluru, S.: Parallel pairwise correlationcomputation on intel xeon phi clusters. arXiv preprint arXiv:1605.01584 (2016)
  17. 17.
    Luk, C.K., Hong, S., Kim, H.: Qilin: exploiting parallelism on heterogeneous multiprocessors with adaptive mapping. In: 42nd Annual IEEE/ACM International Symposium on Microarchitecture, 2009, MICRO-42, pp. 45–55. IEEE (2009)Google Scholar
  18. 18.
    Memeti, S., Pllana, S.: PaREM: a novel approach for parallel regular expression matching. In: 17th International Conference on Computational Science and Engineering (CSE 2014), pp. 690–697, December 2014Google Scholar
  19. 19.
    Memeti, S., Pllana, S.: Accelerating DNA sequence analysis using Intel Xeon Phi. In: PBio at the 2015 IEEE International Symposiumon Parallel and Distributed Processing with Applications (ISPA). IEEE (2015)Google Scholar
  20. 20.
    Memeti, S., Pllana, S.: Analyzing large-scale DNA sequences on multi-core architectures. In: 18th IEEE International Conference on Computational Science and Engineering (CSE 2015). IEEE (2015)Google Scholar
  21. 21.
    Nakao, M., Lee, J., Boku, T., Sato, M.: XcalableMP implementationand performance of NAS parallel benchmarks. In: Proceedings of the Fourth Conference on Partitioned Global Address Space Programming Model, p. 11. ACM (2010)Google Scholar
  22. 22.
    NCBI: National Center for Biotechnology Information U.S. NationalLibrary of Medicine (2015). Accessed Dec 2015
  23. 23.
    Odajima, T., Boku, T., Hanawa, T., Lee, J., Sato, M.: GPU/CPU work sharing with parallel language XcalableMP-dev for parallelized accelerated computing. In: 2012 41st International Conference on Parallel Processing Workshops (ICPPW), pp. 97–106. IEEE (2012)Google Scholar
  24. 24.
    Pllana, S., Benkner, S., Xhafa, F., Barolli, L.: Hybrid performance modeling and prediction of large-scale computing systems. In: International Conference on Complex, Intelligent and Software Intensive Systems, 2008, CISIS 2008, pp. 132–138, March 2008Google Scholar
  25. 25.
    Press, W.H., Teukolsky, S.A., Vetterling, W.T., Flannery, B.P.: Numerical Recipes in C: The Art of Scientific Computing, 3rd edn. Cambridge University Press, Cambridge (2007)zbMATHGoogle Scholar
  26. 26.
    Ravi, V.T., Agrawal, G.: A dynamic scheduling framework for emerging heterogeneous systems. In: 2011 18th International Conference on High Performance Computing (HiPC), pp. 1–10. IEEE (2011)Google Scholar
  27. 27.
    Scogland, T.R.W., Feng, W., Rountree, B., Supinski, B.R.: CoreTSAR: adaptive worksharing for heterogeneous systems. In: Kunkel, J.M., Ludwig, T., Meuer, H.W. (eds.) ISC 2014. LNCS, vol. 8488, pp. 172–186. Springer, Heidelberg (2014). doi: 10.1007/978-3-319-07518-1_11 Google Scholar
  28. 28.
    Viebke, A., Pllana, S.: The potential of the Intel (R) Xeon Phi forsupervised deep learning. In: 2015 IEEE 17th International Conference on High Performance Computing and Communications (HPCC), pp. 758–765. IEEE (2015)Google Scholar

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  1. 1.Department of Computer ScienceLinnaeus UniversityVaxjoSweden

Personalised recommendations