Maximum Performance Computing with Dataflow Engines

  • Oliver Pell
  • Oskar Mencer
  • Kuen Hung Tsoi
  • Wayne Luk


Maximum Performance Computing (MPC) means striving to deliver the maximum possible performance within a space and/or power budget. The essence of the method is to start with a particular application and develop an appropriate computer by iterating between algorithm optimization and machine optimization, essentially, cross-optimizing across the layers of abstraction from mathematics to logic gates. An MPC system pairs fast scalar processors with dataflow engines which can be emulated on FPGAs. In this chapter we outline the general approach, and describe in detail example hardware architecture, programming model and tools. We also discuss additional issues that arise at the cluster level, and describe a detailed case study of applying MPC to Reverse Time Migration, a computational geophysics algorithm widely used in the oil industry.


Message Passing Interface Memory Bandwidth Power Budget Reverse Time Migration Storage Area Network 
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.
    J.B. Dennis, Data flow supercomputers. Computer 13(11), 48–56 (1980)CrossRefGoogle Scholar
  2. 2.
    E. Fermi, J. Pasta, S. Ulam, Studies of nonlinear problems. Tech. Rep. Document LA-1940 (1955)Google Scholar
  3. 3.
    H. Fu, R.G. Clapp, O. Mencer, O. Pell, Accelerating 3D convolution using streaming architectures on FPGAs. SEG Expanded Abstracts 28 (2009)Google Scholar
  4. 4.
    H.T. Kung, Why systolic architectures? Computer 15 (1982)Google Scholar
  5. 5.
    C.E. Leiserson, Fat-trees: universal networks for hardware-efficient supercomputing. IEEE Trans. Comp. 34 (1985)Google Scholar
  6. 6.
    O. Lindtjorn, R.G. Clapp, O. Pell, O. Mencer, M.J. Flynn, H. Fu, Beyond traditional microprocessors for geoscience high-performance computing applications. IEEE Micro 31(2), 41–49 (2011)CrossRefGoogle Scholar
  7. 7.
    O. Lindtjorn, R.G. Clapp, O. Pell, O. Mencer, M.J. Flynn, H. Fu, Beyond traditional microprocessors for geoscience high-performance computing applications. IEEE Micro (2011)Google Scholar
  8. 8.
    W. Liu et al., Anisotropic reverse-time migration using co-processors. SEG Expanded Abstracts 28 (2009)Google Scholar
  9. 9.
    Lustre, Accessed 26th Jaunary 2013
  10. 10.
    Maxeler Technologies: MaxGenFD white paper. Tech. rep. (2010)Google Scholar
  11. 11.
    T. Nemeth, J. Stefani, W. Liu, R. Dimond, O. Pell, R. Ergas, An implementation of the acoustic wave equation on FPGAs. SEG Expanded Abstracts 27 (2008)Google Scholar
  12. 12.
    D. Oriato, O. Pell, C. Andreoletti, N. Bienati, FD modeling beyond 70hz with FPGA acceleration, in SEG 2010 HPC Workshop, Denver (2010)Google Scholar
  13. 13.
    O. Pell, O. Mencer, Surviving the end of frequency scaling with reconfigurable dataflow computing. SIGARCH Comput. Archit. News 39(4), 60–65 (2011)CrossRefGoogle Scholar
  14. 14.
    B. Schroeder, E. Pinheiro, W.D. Weber, DRAM errors in the wild: a large-scale field study, in Proceedings of eleventh International Joint Conference on Measurement and Modeling of Computer Systems (SIGMETRICS ’09) (ACM, New York, USA, 2009), pp. 193–204. DOI 10.1145/1555349.1555372,
  15. 15.
    K.V. Vishwanath, N. Nagappan, Characterizing cloud computing hardware reliability, in Proceedings of the 1st ACM Symposium on Cloud Computing (SoCC ’10) (ACM, New York, USA, 2010), pp. 193–204. DOI 10.1145/1807128.1807161,
  16. 16.
    S. Weston, J. Spooner, S. Racaniere, O. Mencer, Rapid computation of value and risk for derivatives portfolios. Concurrency Comput.: Pract. Ex. (2011)Google Scholar
  17. 17.
    K. Yoon, C. Chin, S. Suh, L. Lines, S. Hong, 3D reverse-time migration using the acoustic wave equation: an experience with the SEG/EAGE dataset. The Leading Edge 22, 38–41 (2003)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2013

Authors and Affiliations

  • Oliver Pell
    • 1
  • Oskar Mencer
    • 1
  • Kuen Hung Tsoi
    • 2
  • Wayne Luk
    • 2
  1. 1.Maxeler TechnologiesLondonUK
  2. 2.Imperial College LondonLondonUK

Personalised recommendations