Autocorrelation analysis: a new and improved method for measuring branch predictability

Branch taken rate and transition rate have been proposed as metrics to characterize the branch predictability. However, these two metrics may misclassify branches with regular history patterns as hard-to-predict branches, causing an inaccurate and ambiguous view of branch predictability. This study uses autocorrelation to analyze the branch history patterns and presents a new metric Degree of Pattern Irregularity (DPI) for branch classification. The proposed metric is evaluated with different branch predictors, and the results show that DPI significantly improves the quality and the accuracy of branch classification over traditional taken rate and transition rate.

[1]  Matthew K. Farrens,et al.  Branch transition rate: a new metric for improved branch classification analysis , 2000, Proceedings Sixth International Symposium on High-Performance Computer Architecture. HPCA-6 (Cat. No.PR00550).

[2]  E. Brigham,et al.  The fast Fourier transform , 2016, IEEE Spectrum.

[3]  Lieven Eeckhout,et al.  Automated microprocessor stressmark generation , 2008, 2008 IEEE 14th International Symposium on High Performance Computer Architecture.

[4]  Yale N. Patt,et al.  Branch classification: a new mechanism for improving branch predictor performance , 1994, MICRO.

[5]  Daniel A. Jiménez,et al.  Dynamic branch prediction with perceptrons , 2001, Proceedings HPCA Seventh International Symposium on High-Performance Computer Architecture.