Modeling the Productivity of HPC Systems on a Computing Center Scale

  • Sandra WienkeEmail author
  • Hristo Iliev
  • Dieter an Mey
  • Matthias S. Müller
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9137)


In pursue of exaflop computing, the expenses of HPC centers increase in terms of acquisition, energy, employment, and programming. Thus, a quantifiable metric for productivity as value per cost gets more important to make an informed decision on how to invest available budgets. In this work, we model overall productivity from a computing center’s perspective. The productivity model uses as value the number of application runs possible during the lifetime of a given supercomputer. The cost is the total cost of ownership (TCO) of an HPC center including costs for administration and programming effort. For the latter, we include techniques for software cost estimation of large codes taken from the domain of software engineering. As tuning effort increases when more performance is required, we further focus on the impact of the 80-20 rule when it comes to development effort. Here, performance can be expressed with respect to Amdahl’s law. Moreover, we include an asymptotic analysis for parameters like number of compute nodes and lifetime. We evaluate our approach on a real-world case: an engineering application in our integrative hosting environment.


Productivity Cost-benefit ratio Cost efficiency TCO Development effort COCOMO 80-20 rule Pareto principle Computing center Scalability 


  1. 1.
    Apon, A., Ahalt, S., Dantuluri, V., Gurdgiev, C., Limayem, M., Ngo, L., Stealey, M.: High performance computing instrumentation and research productivity in US universities. J. Inf. Technol. Impact 10(2), 87–98 (2010)Google Scholar
  2. 2.
    Boehm, B., Abts, C., Brown, A.W., Chulani, S., Clark, B., Horowitz, E., Madachy, R., Reifer, D., Steece, B.: COCOMO II Model Definition Manual, Version 2.1. Technical report, University of Southern California (2000)Google Scholar
  3. 3.
    Chew, W.: No-nonsense guide to measuring productivity. Harvard Bus. Rev. 66(1), 110–118 (1988)MathSciNetGoogle Scholar
  4. 4.
    Culler, D.E., Karp, R.M., Patterson, D.A., Sahay, A., Schauser, K.E., Santos, E., Subramonian, R., von Eicken, T.: LogP: Towards a Realistic Model of Parallel Computation. Technical report, Berkeley (1992)Google Scholar
  5. 5.
    Dongarra, J., Graybill, R., Harrod, W., Lucas, R., Lusk, E., Luszczek, P., Mcmahon, J., Snavely, A., Vetter, J., Yelick, K., Alam, S., Campbell, R., Carrington, L., Chen, T.Y., Khalili, O., Meredith, J., Tikir, M.: DARPA’s HPCS program: history, models, tools, languages. In: Zelkowitz, M.V. (ed.) Advances in COMPUTERS High Performance Computing, Advances in Computers, vol. 72, pp. 1–100. Elsevier (2008)Google Scholar
  6. 6.
    Ebcioglu, K., Sarkar, V., El-Ghazawi, T., Urbanic, J., Center, P.: An experiment in measuring the productivity of three parallel programming languages. In: Workshop on Productivity and Performance in High-End Computing (P-PHEC), pp. 30–36 (2006)Google Scholar
  7. 7.
    European Commission: Guide to Financial Issues relating to FP7 Indirect Actions (2013)Google Scholar
  8. 8.
    Faulk, S., Gustafson, J., Johnson, P., Porter, A., Tichy, W., Votta, L.: Measuring high performance computing productivity. Int. J. High Perform. Comput. Appl. 18(4), 459–473 (2004)CrossRefGoogle Scholar
  9. 9.
    German Science Foundation (DFG): Personalmittelsätze der DFG für das Jahr (2013)Google Scholar
  10. 10.
  11. 11.
    InfiniBand Trade Association (2015).
  12. 12.
    Intel Corporation: Intel Processor ARK (2015).
  13. 13.
    Joseph, E.C., Conway, S., Dekate, C.: Creating Economic Models Showing the Relationship Between Investments in HPC and the Resulting Financial ROI and Innovation and How It Can Impact a Nation’s Competitiveness and Innovation. International Data Corporation (IDC), Technical report (2013)Google Scholar
  14. 14.
    Kennedy, K., Koelbel, C., Schreiber, R.: Defining and measuring the productivity of programming languages. Int. J. High Perform. Comput. Appl. 18(4), 441–448 (2004)CrossRefGoogle Scholar
  15. 15.
    Kepner, J.: High performance computing productivity model synthesis. Int. J. High Perform. Comput. Appl. 18(4), 505–516 (2004)CrossRefGoogle Scholar
  16. 16.
    McConnell, S.: Software Estimation: Demystifying the Black Art. Redmond, Wa. Microsoft Press (2006)Google Scholar
  17. 17.
    McCracken, M., Wolter, N., Snavely, A.: Beyond performance tools: Measuring and modeling productivity in HPC. In: Third International Workshop on Software Engineering for High Performance Computing Applications, SE-HPC 2007, pp. 4–4 (2007)Google Scholar
  18. 18.
    Murphy, D., Nash, T., Lawrence Votta, J., Kepner, J.: A System-wide Productivity Figure of Merit. CT Watch Quarterly 2(4B) (2006)Google Scholar
  19. 19.
    Newman, M.: Power laws, Pareto distributions and Zipf’s law. Contemp. Phys. 46(5), 323–351 (2005)CrossRefGoogle Scholar
  20. 20.
    Pekurovsky, D.: P3DFFT: a framework for parallel computations of Fourier transforms in three dimensions. SIAM J. Sci. Comput. 34(4), C192–C209 (2012)zbMATHMathSciNetCrossRefGoogle Scholar
  21. 21.
    Sadowski, C., Shewmaker, A.: The Last Mile: Parallel Programming and Usability. In: Proceedings of the FSE/SDP Workshop on Future of Software Engineering Research, FoSER 2010, pp. 309–314. ACM, New York (2010)Google Scholar
  22. 22.
    Snir, M., Bader, D.A.: A framework for measuring supercomputer productivity. Int. J. High Perform. Comput. Appl. 18(4), 417–432 (2004)CrossRefGoogle Scholar
  23. 23.
    Sterling, T.: Productivity metrics and models for high performance computing. Int. J. High Perform. Comput. Appl. 18(4), 433–440 (2004)MathSciNetCrossRefGoogle Scholar
  24. 24.
    Wang, L., Khan, S.: Review of performance metrics for green data centers: a taxonomy study. J. Supercomputing 63(3), 639–656 (2011)CrossRefGoogle Scholar
  25. 25.
    Wienke, S., Iliev, H., Hahnfeld, J., an Mey, D., Müller, M.S.: \(aixH(PC)^2\) - Aachen HPC Productivity Calculator (2015).
  26. 26.
    Wienke, S., an Mey, D., Müller, M.S.: Accelerators for technical computing: is it worth the pain? A TCO perspective. In: Kunkel, J.M., Ludwig, T., Meuer, H.W. (eds.) ISC 2013. LNCS, vol. 7905, pp. 330–342. Springer, Heidelberg (2013) CrossRefGoogle Scholar
  27. 27.
    Williams, S., Waterman, A., Patterson, D.: Roofline: an insightful visual performance model for multicore architectures. Commun. ACM 52(4), 65–76 (2009)CrossRefGoogle Scholar
  28. 28.
    Zelkowitz, M., Basili, V., Asgari, S., Hochstein, L., Hollingsworth, J., Nakamura, T.: Measuring productivity on high performance computers. In: IEEE International Symposium on Software Metrics, p. 6 (2005).
  29. 29.
    Zelkowitz, M., Hollingsworth, J., Basili, V., Asgari, S., Shull, F., Carver, J., Hochstein, L.: Parallel Programmer Productivity: A Case Study of Novice Parallel Programmers. SC Conference 35 (2005).

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • Sandra Wienke
    • 1
    • 2
    • 3
    Email author
  • Hristo Iliev
    • 1
    • 2
    • 3
  • Dieter an Mey
    • 1
    • 2
    • 3
  • Matthias S. Müller
    • 1
    • 2
    • 3
  1. 1.IT CenterRWTH Aachen UniversityAachenGermany
  2. 2.Chair for High Performance ComputingRWTH Aachen UniversityAachenGermany
  3. 3.JARA – High-Performance ComputingAachenGermany

Personalised recommendations