Peers-for-peers (P4P): an efficient and reliable fault-tolerance strategy for cycle-stealing P2P applications
by Ankur Gupta, Lalit K. Awasthi
International Journal of Communication Networks and Distributed Systems (IJCNDS), Vol. 6, No. 2, 2011

Abstract: Cycle-stealing P2P applications need to contend with a transient node population, which requires specialised approaches to be formulated to achieve fault-tolerance. This research paper proposes the peers-for-peers (P4P), an efficient and reliable strategy, based on selecting suitable peers to provide fault-tolerance for P2P applications by redundantly storing check-pointing or message logging data, ensuring its survivability. The strategy is suitable for cycle-stealing P2P applications and is optimised to reduce the message exchange overheads compared to other existing approaches, while alleviating the need to retrieve/reconstruct check-pointing data thereby improving application performance. Simulation results establish the effectiveness of the strategy in choosing quality nodes which meet the application computing requirements and ensuring survivability of check-pointing data in the face of heavy node transience. A real-world implementation based on the JXTA framework establishes its applicability to cycle-stealing P2P applications.

Online publication date: Thu, 26-Feb-2015

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Communication Networks and Distributed Systems (IJCNDS):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com