Advertisement

Understanding Prediction Limits Through Unbiased Branches

  • Lucian Vintan
  • Arpad Gellert
  • Adrian Florea
  • Marius Oancea
  • Colin Egan
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4186)

Abstract

The majority of currently available branch predictors base their prediction accuracy on the previous k branch outcomes. Such predictors sustain high prediction accuracy but they do not consider the impact of unbiased branches which are difficult-to-predict. In this paper, we quantify and evaluate the impact of unbiased branches and show that any gain in prediction accuracy is proportional to the frequency of unbiased branches. By using the SPECcpu2000 integer benchmarks we show that there are a significant proportion of unbiased branches which severely impact on prediction accuracy (averaging between 6% and 24% depending on the prediction context used).

Keywords

Prediction Accuracy Distribution Index Polarisation Index High Prediction Accuracy Path Information 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Burger, D., Goodman, J.R.: Billion Transistor Architectures. IEEE Computer, 46–49 (September 1997)Google Scholar
  2. 2.
    Chang, P.-Y., Hao, E., Yeh, T.-Y., Patt, Y.N.: Branch Classification: a New Mechanism for Improving Branch Predictor Performance. In: Proceedings of the 27th International Symposium on Microarchitecture, San Jose, California (1994)Google Scholar
  3. 3.
    Chappell, R., Tseng, F., Yoaz, A., Patt, Y.: Difficult-Path Branch Prediction Using Subordinate Microthreads. In: The 29th Annual International Symposia on Computer Architecture, Alaska, USA (May 2002)Google Scholar
  4. 4.
    Desmet, V., Eeckhout, L., De Bosschere, K.: Evaluation of the Gini-index for Studying Branch Prediction Features. In: Proceedings of the 6th International Conference on Computing Anticipatory Systems (CASYS). AIP Conference Proceedings, vol. 718, pp. 376–384 (2004)Google Scholar
  5. 5.
    Hennessy, J., Patterson, D.: Computer Architecture: A Quantitative Approach, 3rd edn. Morgan Kaufmann Publishers, San Francisco (2003)Google Scholar
  6. 6.
    Jiménez, D.A., Lin, C.: Dynamic Branch Prediction with Perceptrons. In: Proceedings of the 7th International Symposium on High Performance Computer Architecture (January 2001)Google Scholar
  7. 7.
    Loh, G.H.: Simulation Differences Between Academia and Industry: A Branch Prediction Case Study. In: International Symposium on Performance Analysis of Software and Systems (ISPASS), Austin, TX, USA, pp. 21–31 (March 2005)Google Scholar
  8. 8.
    McFarling, S.: Combining Branch Predictors, WRL Technical Note TN-36, Digital Equipment Corporation (June 1993)Google Scholar
  9. 9.
    Pan, S.T., So, K., Rahmeh, J.T.: Improving the accuracy of dynamic branch prediction using branch correlation. In: Proceedings of ASPLOS V, Boston, MA, pp. 76–84 (October 1992)Google Scholar
  10. 10.
    Patt, Y.N., Patel, S.J., Friendly, D.H., Stark, J.: One Billion Transistors, One Uniprocessor, One Chip. IEEE Computer 1, 51–57 (1997)Google Scholar
  11. 11.
    Simplescalar The SimpleSim Tool Set, ftp://ftp.cs.wisc.edu/pub/sohi/Code/simplescalar
  12. 12.
    SPEC, The SPEC benchmark programs, http://www.spec.org
  13. 13.
    Yeh, T.Y., Patt, Y.N.: Two-level adaptive branch prediction. In: Proceedings of the 24-the ACM/IEEE International Symposium on Microarchitecture (November 1991)Google Scholar
  14. 14.
    Vintan, L., Egan, C.: Extending Correlation in Branch Prediction Schemes. In: International Euromicro 1999 Conference, Italy (September 1999)Google Scholar
  15. 15.
    Vintan, L., Iridon, M.: Towards a High Performance Neural Branch Predictor. In: International Joint Conference on Neural Networks, Washington DC, USA (July 1999)Google Scholar
  16. 16.
    The 1st JILP Championship Branch Prediction Competition (CBP-1) (2004), http://www.jilp.org/cbp

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Lucian Vintan
    • 1
  • Arpad Gellert
    • 1
  • Adrian Florea
    • 1
  • Marius Oancea
    • 1
  • Colin Egan
    • 2
  1. 1.Computer Science Department“Lucian Blaga” University of SibiuSibiuRomania
  2. 2.School of Computer ScienceUniversity of HertfordshireHatfield, College LaneUK

Personalised recommendations