Alternative implementations of hybrid branch predictors
- 1 November 1995
- proceedings article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
Very accurate branch prediction is an important requirement for achieving high performance on deeply pipelined, superscalar processors. To improve on the prediction accuracy of current single-scheme branch predictors, hybrid (multiple-scheme) branch predictors have been proposed (McFarling (1993), Chang et al. (1994)). These predictors combined multiple single-scheme predictors into a single predictor. They use a selection mechanism to decide for each branch, which single-scheme predictor to use. The performance of a hybrid predictor depends on its single-scheme predictor components and its selection mechanism. Using known single-scheme predictors and selection mechanisms, this paper identifies the most effective hybrid predictor implementation. In addition, it introduces a new selection mechanism, the 2-level selector, which further improves the performance of the hybrid branch predictorKeywords
This publication has 9 references indexed in Scilit:
- A Comparison Of Dynamic Branch Predictors That Use Two Levels Of Branch HistoryPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- A study of branch prediction strategiesPublished by Association for Computing Machinery (ACM) ,1998
- A comparative analysis of schemes for correlated branch predictionPublished by Association for Computing Machinery (ACM) ,1995
- Branch classificationPublished by Association for Computing Machinery (ACM) ,1994
- Predicting conditional branch directions from previous runs of a programPublished by Association for Computing Machinery (ACM) ,1992
- Improving the accuracy of dynamic branch prediction using branch correlationPublished by Association for Computing Machinery (ACM) ,1992
- Alternative implementations of two-level adaptive branch predictionPublished by Association for Computing Machinery (ACM) ,1992
- Two-level adaptive training branch predictionPublished by Association for Computing Machinery (ACM) ,1991
- Reducing the cost of branchesACM SIGARCH Computer Architecture News, 1986