Enter a sentence, or cut and paste a paragraph

Refine

Refine

Refine to these subject areas:

  • Select All | Select None

Advanced Filters »

Advanced Filters

Refine to these subject areas:

  • Select All | Select None

Advanced Filters «

  • Find information that...

  • to

  • Specific Data Sources:

    All Edit

    Select All  |  Select None

Reset filters

Bookmark

A case for two-level distributed recovery schemes

To view this document, you will need to have Adobe Flash Player installed.
Please click here to install.

Most distributed and multiprocessor recovery schemes proposed in the literature are designed to tolerate arbitrary number of failures. In this paper, we demonstrate that, it is often advantageous to use "two-level" recovery schemes. A two-level recovery scheme tolerates the more probable failures with low performance overhead, while the less probable failures may be tolerated with a higher overhead. By minimizing the overhead for the more frequently occurring failure scenarios, our approach is expected to achieve lower performance overhead (on average) as compared to existing recovery schemes.To demonstrate the advantages of two-level recovery, we evaluate the performance of a recovery scheme that takes two different types of checkpoints, namely, 1-checkpoints and N -checkpoints. A single failure can be tolerated by rolling the system back to a 1-checkpoint, while multiple failure recovery is possible by rolling back to an N -checkpoint. For such a system, we demonstrate that to minimize the average overhead, it is often necessary to take both 1-checkpoints and N -checkpoints.While the conclusions of this paper are intuitive, the work on design of appropriate recovery schemes is lacking. The objective of this paper is to motivate research into recovery schemes that can provide multiple levels of fault tolerance.

This is a preview. The total pages displayed will be limited.

Rent for $0.99 FREE

Login

It seems like you have an account, please login to rent this article

Forgot your password?

Don't have an account yet? Sign up now!

To view the full-text of this article, sign up for a free DeepDyve account below.

A free Basic account also comes with
5 free rentals to help get you started.

It seems like you have an account, please login to rent this article

Learn more Existing user? Login here

Article Details
Vaidya, Nitin H.
ACM SIGMETRICS Performance Evaluation Review , Volume 23 (1)
Association for Computing Machinery – May 1, 1995
More Info

More Like This Article

View All dataSource[]=aspet&dataSource[]=aacc&dataSource[]=aacr&dataSource[]=aip&dataSource[]=appi_book&dataSource[]=appi_journal&dataSource[]=asip&dataSource[]=asm&dataSource[]=asn&dataSource[]=annual_reviews&dataSource[]=arxiv&dataSource[]=acm&dataSource[]=clinical_trials&dataSource[]=dailymed&dataSource[]=degruyter&dataSource[]=emea&dataSource[]=epo&dataSource[]=faseb&dataSource[]=gsa&dataSource[]=health_affairs&dataSource[]=hindawi&dataSource[]=iucr&dataSource[]=jbjs&dataSource[]=mary_ann_liebert&dataSource[]=medline&dataSource[]=mit_press&dataSource[]=oxford&dataSource[]=pnas&dataSource[]=psyc_books&dataSource[]=psyc_critiques&dataSource[]=psyc_articles&dataSource[]=plos_journal&dataSource[]=pubmed_central&dataSource[]=sage&dataSource[]=spie&dataSource[]=springer&dataSource[]=taylor_francis&dataSource[]=aps&dataSource[]=the_scientist&dataSource[]=uc_press&dataSource[]=uspto_abstract&dataSource[]=pct

Browse: Journals | Publishers

Bookmark an Article

To bookmark an article, please log in first, or sign up for a DeepDyve account if you don't already have one.

OK

Preview-Only Articles

"Preview Only" refers to those articles which can only be previewed but not rented because DeepDyve does not currently have permission from the publisher. If you would like to download and purchase Preview-Only articles, or any of our Premium articles, you may click the "Download Article from Publisher" link. DeepDyve will take you directly to the article at the publisher's website to complete your transaction.

OK