HUSCAP logo Hokkaido Univ. logo

Hokkaido University Collection of Scholarly and Academic Papers >
情報科学研究科  >
雑誌発表論文等  >

An energy-efficient dynamic branch predictor with a two-clock-cycle naive Bayes classifier for pipelined RISC microprocessors

フルテキスト
8_235.pdf978.6 kBPDF見る/開く
この文献へのリンクには次のURLを使用してください:http://hdl.handle.net/2115/68659

タイトル: An energy-efficient dynamic branch predictor with a two-clock-cycle naive Bayes classifier for pipelined RISC microprocessors
著者: Hida, Itaru 著作を一覧する
Takamaeda-Yamazaki, Shinya 著作を一覧する
Ikebe, Masayuki 著作を一覧する
Motomura, Masato 著作を一覧する
Asai, Tetsuya 著作を一覧する
キーワード: dynamic branch prediction
supervised machine learning
naive Bayes classifier
energy-efficient microprocessor
low-power architecture
CMOS digital circuit
発行日: 2017年
出版者: 電子情報通信学会(The Institute of Electronics, Information and Communication Engineers / IEICE)
誌名: Nonlinear Theory and Its Applications, IEICE
巻: 8
号: 3
開始ページ: 235
終了ページ: 245
出版社 DOI: 10.1587/nolta.8.235
抄録: In this paper, we propose a Bayesian branch-prediction circuit, consisting of an instruction-feature extractor and a naive Bayes classifier (NBC), as a machine learning approach for branch prediction. A branch predictor predicts the outcome of a branch instruction by analyzing the pattern of the previous branch outcome. In other words, branch prediction can be viewed as a type of pattern recognition problem, and such problems are often solved using neural networks. A perceptron branch predictor has already been proposed as one example of a neural branch prediction architecture, which predicts the next branch outcome by using past branch history to form feature vectors. The proposed circuit is constructed by replacing the arithmetic unit (neurons) in conventional neural branch predictors with an NBC. By introducing an approximate Bayesian computation and its parallel architectures, the NBC circuit completes branch prediction within two clock cycles per instruction. This constitutes a suitable replacement for conventional branch predictors in modern pipelined reduced instruction set computing microprocessors.
Rights: Copyright ©2017 The Institute of Electronics, Information and Communication Engineers
資料タイプ: article
URI: http://hdl.handle.net/2115/68659
出現コレクション:雑誌発表論文等 (Peer-reviewed Journal Articles, etc)

提供者: 肥田 格

 

本サイトに関するご意見・お問い合わせは repo at lib.hokudai.ac.jp へお願いします。 - 北海道大学