Branch Classification: A New Mechanism for Improving Branch Predictor Performance

Chang, Po-Yung; Hao, Eric; Yeh, Tse-Yu; Patt, Yale N.

doi:10.1007/bf03356745

Cited by 24 publications

(50 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Branches vary widely in their dynamic behavior, and predictors that work well on one type of branches may not work as well on others. A set of hard-to-predict branches that comprise a fundamental limit to traditional branch predictors can always be identified [17]. We assume that there are two classes: easy-to-predict and hard-to-predict branches, and the expected branch prediction accuracy is higher for the first, and lower for the second.…”

Section: Model Definitionmentioning

confidence: 99%

Fluid Stochastic Petri Nets: From Fluid Atoms in ILP Processor Pipelines to Fluid Atoms in P2P Streaming Networks

Mitrevski¹,

Kotevski²

2012

Petri Nets - Manufacturing and Computer Science

View full text Add to dashboard Cite

Section: Model Definitionmentioning

confidence: 99%

Fluid Stochastic Petri Nets: From Fluid Atoms in ILP Processor Pipelines to Fluid Atoms in P2P Streaming Networks

Mitrevski¹,

Kotevski²

2012

Petri Nets - Manufacturing and Computer Science

View full text Add to dashboard Cite

“…Using this information, the selector associates a currently predicted branch to either global or per-address predictor which is most likely to be more accurate. Using similar amount of hardware resources, this hybrid approach is proved to be more accurate than the stand-alone predictors [4][5][6].…”

Section: Characterization Of Conditionalmentioning

confidence: 99%

“…Many sophisticated conditional branch prediction schemes have been proposed, including twolevel adaptive predictors [2,3], gshare predictor [4], and hybrid predictors [4,5]. These predictors and their variants exploit past history of the predicted branch, history of recent branches, and other branch correlations to accurately predict conditional branches [6,7].…”

Section: Introductionmentioning

confidence: 99%

Improving Conditional Branch Prediction on Speculative Multithreading Architectures

Iwama

Barli

Sakai

et al. 2001

Euro-Par 2001 Parallel Processing

View full text Add to dashboard Cite

Dynamic branch prediction is an indispensable technique for increasing performance in modern processors, and has been actively investigated in the last decade. However, those methods are not readily applicable to speculative multithreading architectures. In this paper, we discuss problems that emerge when conventional conditional branch predictions are used on speculative multithreading architectures, and propose a prediction scheme to improve the prediction accuracy. Evaluation results show that by using our scheme, the average of prediction accuracy of SPEC95int applications can be improved from 89.7% to 90.4%, approaching the accuracy on single threaded execution of 92.9%.

show abstract

“…In microrocessors, the accuracy of the branch prediction is surprisingly high -more than 95 % are reported in (Chang et al, 1994) for single SPEC benchmark programs. However, rerolling execution in case of a wrongly predicted path is costly in terms of processor cycles, especially in deeply pipelined microprocessors.…”

Section: Micro Dataflowmentioning

confidence: 99%

Beyond Dataflow

Robi�¹,

�ilc²,

Ungerer³

2000

CIT

View full text Add to dashboard Cite

This paper presents some recent advanced dataflow architectures. While the dataflow concept offers the potential of high performance, the performance of an actual dataflow implementation can be restricted by a limited number of functional units, limited memory bandwidth, and the need to associatively match pending operations with available functional units. Since the early 1970s, there have been significant developments in both fundamental research and practical realizations of dataflow models of computation. In particular, there has been active research and development in multithreaded architectures that evolved from the dataflow model. Also some other techniques for combining control-flow and dataflow emerged, such as coarse-grain dataflow, dataflow with complex machine operations, RISC dataflow, and micro dataflow. These developments have also had certain impact on the conception of highperformance superscalar processors in the "post-RISC" era.

show abstract

Branch Classification: A New Mechanism for Improving Branch Predictor Performance

Cited by 24 publications

References 10 publications

Fluid Stochastic Petri Nets: From Fluid Atoms in ILP Processor Pipelines to Fluid Atoms in P2P Streaming Networks

Fluid Stochastic Petri Nets: From Fluid Atoms in ILP Processor Pipelines to Fluid Atoms in P2P Streaming Networks

Improving Conditional Branch Prediction on Speculative Multithreading Architectures

Beyond Dataflow

Contact Info

Product

Resources

About