期刊名称：International Journal of Computer Science & Technology
印刷版ISSN：2229-4333
电子版ISSN：0976-8491
出版年度：2013
卷号：4
期号：3
页码：59-65
语种：English
出版社：Ayushmaan Technologies
摘要：Data dependencies between instructions have traditionally limited the ability of processors to execute instructions in parallel. Data value predictors are used to overcome these dependencies by guessing the outcomes of instruction in a program. Because mispredictions can result in a significant performance decrease, most data value predictors include a confidence estimator that indicates whether a prediction should be used or not. Much research has been done recently in the area of data value prediction as a means of overcoming these data dependencies [7-11, 17-18, 20-21]. The goal of data value prediction is to guess the outcome of an instruction before the instruction is actually executed, allowing future instructions that depend on its outcome to be executed sooner. This paper presents a global approach to confidence estimation in which the prediction accuracy of previous instructions is used to estimate the confidence of the current prediction. Data value prediction is done using data value predictors. Support Vector Machines are used to identify which past instructions affect the accuracy of a prediction and to decide based on their results whether the prediction is likely to be correct or not.
