Potentials and Limitations of Fault-Based Markov Prefetching for Virtual Memory Pages Gretta Batiels, Anna Karlin, Henry Levy, Geoffrey Voelker Department of Computer Science and Engineering University of Washington Seattle, WA Darrell Anderson, Jeffrey Chase Department of Computer Science Duke University Durham, NC {gretta,karlin,levy,voelker}@cs.washington.edu {anderson,chase}@cs.duke.edu potential benefit of successful prefetching. Our goal is to examine this tradeoff to determine the potential for speculative prefetching in this high-speednetwork environment. 1. INTRODUCTION Prefetching non-resident pages into memory before they are accessed greatly reduce I/O stall time when comparedwith a can fetch-on-demand strategy. While much recent research has focused on prefetching algorithms, some of this work assumes complete knowledge of a program s future referencestream[3,4]. Without such future knowledge, however, two limitations severely restrict the effectiveness of prefetching. First, with the exception of sequentially accessedfiles, prediction of future accesspatterns is difficult. Second, given the large latency of disk transfers,the cost of prediction errors is high, and therefore prefetching from disk may actually degrade performance by wasting critical systemresources. This work focuses on fault-based predictive prefetching in highspeedlocal-areanetworks. Our approachis fault-basedbecauseit accumulates access information only at page-fault events; the advantageof this schemeis that it is highly efficient and easily implementable within the operating
End of preview. The entire article is 2 pages. To view the full-text, please rent this article to continue.
/lp/association-for-computing-machinery/potentials-and-limitations-of-fault-based-markov-prefetching-for-HI2c76t0mu
Welcome to DeepDyve! Rent Premier Research Articles and Save Up to 90%
Potentials and limitations of fault-based Markov prefetching for virtual memory pages
Bartels, Gretta; Karlin, Anna; Anderson, Darrell; Chase, Jeffrey; Levy, Henry; Voelker, Geoffrey
Follow ACM SIGMETRICS Performance Evaluation Review
, Volume 27 (1)
Association for Computing Machinery – Jun 1, 1999
More Info
More Like This Article
View All dataSource[]=actageo&dataSource[]=aspet&dataSource[]=aaos&dataSource[]=aacc&dataSource[]=aacr&dataSource[]=aea&dataSource[]=aip&dataSource[]=ajnr&dataSource[]=ams&dataSource[]=aps_physical&dataSource[]=appi_book&dataSource[]=appi_journal&dataSource[]=apha&dataSource[]=asip&dataSource[]=asm&dataSource[]=asn&dataSource[]=aspb&dataSource[]=avs&dataSource[]=annual_reviews&dataSource[]=arxiv&dataSource[]=acm&dataSource[]=berghahn&dataSource[]=cabi&dataSource[]=clinical_trials&dataSource[]=dailymed&dataSource[]=degruyter&dataSource[]=du_press&dataSource[]=esa&dataSource[]=eu_press&dataSource[]=elsevier&dataSource[]=emerald&dataSource[]=ejtr&dataSource[]=emea&dataSource[]=epo&dataSource[]=faseb&dataSource[]=gsa&dataSource[]=health_affairs&dataSource[]=hindawi&dataSource[]=imanager&dataSource[]=imedpub&dataSource[]=informa_healthcare&dataSource[]=informs&dataSource[]=iop&dataSource[]=iucr&dataSource[]=iospress&dataSource[]=jbjs&dataSource[]=leftcoast&dataSource[]=lu_press&dataSource[]=mesharpe&dataSource[]=mary_ann_liebert&dataSource[]=medline&dataSource[]=mit_press&dataSource[]=nature&dataSource[]=oxford&dataSource[]=pier_professional&dataSource[]=pnas&dataSource[]=portlandpress&dataSource[]=psyc_articles&dataSource[]=psyc_books&dataSource[]=psyc_critiques&dataSource[]=plos_journal&dataSource[]=pubmed_central&dataSource[]=rsna&dataSource[]=rockefeller&dataSource[]=rcn&dataSource[]=ria&dataSource[]=rsc&dataSource[]=sage&dataSource[]=spie&dataSource[]=springer_journal&dataSource[]=springer&dataSource[]=taylor_francis&dataSource[]=aps&dataSource[]=the_scientist&dataSource[]=uc_press&dataSource[]=uspto_abstract&dataSource[]=wiley&dataSource[]=pct
© 2012 DeepDyve, Inc. All rights reserved.
Terms of Service | Privacy Policy
