FiNS: A Framework for Accelerating Nested Simulations on Heterogeneous Platforms

  • Joris CramwinckelEmail author
  • Stefan Singor
  • Ana Lucia Varbanescu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9523)


Insurers use advanced risk management models to, among other things, compute required capital for different sources of financial risks. In these models the application of nested simulations becomes increasingly important. To keep computation times within acceptable limits high-performance computing is required. In this work we present a framework designed to significantly improve the performance of nested simulations by using heterogeneous computing. Specifically, we use modern features from CUDA - streams, Hyper-Q, and Multi-Process Service - to take full advantage of the massive parallelism of modern GPUs. We manage to reduce the execution time of such simulations from several hours to tens of minutes.


CPU-GPU heterogeneous computing Asset & Liability Management Nested simulations CUDA Streams CUDA Hyper-Q 


  1. 1.
    Abbas-Turki, L., Vialle, S., Lapeyre, B., Mercier, P.: Pricing derivatives on graphics processing units using monte carlo simulation. Concurr. Comput.: Pract. Exp. 26(9), 1679–1697 (2014). CrossRefGoogle Scholar
  2. 2.
    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
  3. 3.
    Bauer, D., Reuss, A., Singer, D.: On the calculation of the solvency capital requirement based on nested simulations. Astin Bull. 42(02), 453–499 (2012)zbMATHMathSciNetGoogle Scholar
  4. 4.
    Bernemann, A., Schreyer, R., Spanderen, K.: Accelerating exotic option pricing and model calibration using gpus (2011).
  5. 5.
    Cramwinckel, J.: GPU Accelerated framework for financial nested simulations. Master’s thesis. VU University Amsterdam (2015)Google Scholar
  6. 6.
    Fernández, J., Ferreiro, A., García-Rodríguez, J., Leitao, A., López-Salas, J., Vázquez, C.: Static and dynamic sabr stochastic volatility models: calibration and option pricing using gpus. Math. Comput. Simul. 94, 55–75 (2013)CrossRefGoogle Scholar
  7. 7.
    Glasserman, P.: Monte Carlo Methods in Financial Engineering, vol. 53. Springer, New York (2004)zbMATHGoogle Scholar
  8. 8.
    Joshi, M.S.: Graphical asian options. Wilmott J. 2(2), 97–107 (2010)CrossRefGoogle Scholar
  9. 9.
    Kofler, K., Grasso, I., Cosenza, B., Fahringer, T.: An automatic input-sensitive approach for heterogeneous task partitioning. In: ICS 2013 (2013)Google Scholar
  10. 10.
    Lee, A., Yau, C., Giles, M.B., Doucet, A., Holmes, C.C.: On the utility of graphics cards to perform massively parallel simulation of advanced monte carlo methods. J. Comput. Graph. Stat. 19(4), 769–789 (2010)CrossRefGoogle Scholar
  11. 11.
    Luitjens, J.: CUDA streams - best practices and pitfalls, Presentation GTC 2014 (2014)Google Scholar
  12. 12.
    Luk, C.K., Hong, S., Kim, H.: Qilin: exploiting parallelism on heterogeneous multiprocessors with adaptive mapping. In: MICRO 2009 (2009)Google Scholar
  13. 13.
    NVIDIA: Kepler GK110 architecture whitepaper, v1.0 (2012)Google Scholar
  14. 14.
    NVIDIA: Sharing a GPU between MPI processes: Multi-process service (MPS) overview (2013). Technical Brief TB-06737-003Google Scholar
  15. 15.
    Planas, J., Badia, R.M., Ayguadé, E., Labarta, J.: Self-Adaptive ompss tasks in heterogeneous environments. In: IPDPS 2013 (2013)Google Scholar
  16. 16.
    Shen, J., Varbanescu, A.L., Sips, H.: Look before you leap: using the right hardware resources to accelerate applications. In: HPCC (2014)Google Scholar
  17. 17.
    Shen, J., Varbanescu, A.L., Zou, P., Lu, Y., Sips, H.: Improving performance by matching imbalanced workloads with heterogeneous platforms. In: ICS (2014)Google Scholar
  18. 18.
    Tian, Y., Zhu, Z., Klebaner, F.C., Hamza, K.: Option pricing with the sabr model on the gpu. In: 2010 IEEE Workshop on High Performance Computational Finance (WHPCF). IEEE (2010)Google Scholar
  19. 19.
    Tian, Y., Zhu, Z., Klebaner, F.C., Hamza, K.: Pricing barrier and american options under the sabr model on the graphics processing unit. Concurr. Comput.: Pract. Exp. 24(8), 867–879 (2012)CrossRefGoogle Scholar
  20. 20.
    Wende, F., Steinke, T., Cordes, F.: Multi-threaded kernel offloading to gpgpu using hyper-q on kepler architecture. ZIB-Rep. 14–19 June 2014 (2014)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • Joris Cramwinckel
    • 1
    Email author
  • Stefan Singor
    • 1
    • 2
  • Ana Lucia Varbanescu
    • 3
  1. 1.Ortec FinanceRotterdamThe Netherlands
  2. 2.Delft University of TechnologyDelftThe Netherlands
  3. 3.University of AmsterdamAmsterdamThe Netherlands

Personalised recommendations