Transactions on High-Performance Embedded Architectures and Compilers II (Paperback)

David Whalley

ISBN 10: 3642009034 / ISBN 13: 9783642009037

Published by Springer-Verlag Berlin and Heidelberg GmbH & Co. KG, Berlin, 2009

Paperback. 1 2 Per Stenstro .m and David Whalley 1 Chalmers University of Technology, Sweden 2 Florida State University, U.S.A. In January2007,the secondedition in the series of International Conferenceson High-Performance Embedded Architectures andCompilers (HiPEAC'2007)was held in Ghent,Belgium.We were fortunate to attract around70 submissions of whichonly19wereselected forpresentation.Amongthese,weaskedtheauthors ofthe?vemost highly rated contributionsto make extended versions ofthem. They all accepted to do that andtheirarticles appear in this section ofthe second volume. The?rstarticlebyKeramidas,Xekalakis,andKaxirasfocusesontheincreased power consumption in set-associativecaches.They presenta novel approach to reduce dynamicpower that leverages on the previously proposed cache decay approach that has been shown to reduce static (or leakage) power. In the secondarticlebyMagarajan,Gupta,andKrishnaswamythe focus ison techniques to encrypt data in memory to preservedata integrity. The problem with previous techniques is that the decryption latency ends up on the critical memory access path. Especially in embedded processors,caches are small and it isdi?cultto hide the decryption latency.The authors propose a compiler-based strategy that manages to reduce the impact of the decryption time signi?cantly. The thirdarticlebyKluyskensandEeckhoutfocusesondetailedarchitectural simulation techniques.It is well-known that they are ine?cientandaremedy to the problem isto use sampling.When usingsampling,onehastowarm up memory structures such as caches andbranch predictors.Thispaper introduces a noveltechnique calledBranchHistoryMatchingfore?cient warmupofbranch predictors. The fourth articlebyBhadauria,McKee,Singh, and Tyson focuses on static power consumptioninlarge caches.Theyintroduce a reuse-distance drowsy cache mechanism that issimpleas well as e?ective in reducingthestaticpower in caches. This book contains extended versions of key papers from the 2nd International Conference on High-Performance Embedded Architectures and Compilers (HiPEAC 2007). It also covers such topics as microarchitecture, code generation, and performance modeling. Shipping may be from multiple locations in the US or from the UK, depending on stock availability. Seller Inventory # 9783642009037

Synopsis:

caches.Theyintroduce a reuse-distance drowsy cache mechanism that issimpleas well as e?ective in reducingthestaticpower in caches.

From the Back Cover:

Transactions on HiPEAC aims at the timely dissemination of research contributions in computer architecture and compilation methods for high-performance embedded computer systems. Recognizing the convergence of embedded and general-purpose computer systems, this journal publishes original research on systems targeted at specific computing tasks as well as systems with broad application bases. The scope of the journal therefore covers all aspects of computer architecture, code generation and compiler optimization methods of interest to researchers and practitioners designing future embedded systems.

This second issue contains 15 papers carefully reviewed and selected out of 31 submissions and is divided into two sections. The first section contains extended versions of the top five papers from the 2nd International Conference on High-Performance Embedded Architectures and Compilers (HiPEAC 2007) held in Ghent, Belgium, in January 2007.  The second section consists of ten papers covering topics such as microarchitecture, memory systems, code generation, and performance modeling.

Bibliographic Details

Title: Transactions on High-Performance Embedded ...
Publisher: Springer-Verlag Berlin and Heidelberg GmbH & Co. KG, Berlin
Publication Date: 2009
Binding: Paperback
Condition: new