UTHM Institutional Repository

On affirmative adaptive failure detection

Mohd Noor, Ahmad Shukri and Mat Deris, Mustafa and Herawan, Tutut and Nor Hassan, Mohamad (2012) On affirmative adaptive failure detection. In: Proceedings of the 12th international conference on Algorithms and Architectures for Parallel Processing (ICA3PP'12), 4-7 September 2012, Fukuoka, Japan.

Full text not available from this repository.

Abstract

Fault detection methodology is a crucial part in providing a scalable, dependable and high availability of grid computing environment. The most popular technique that used in detecting fault is heartbeat mechanism where it monitors the grid resources in a very short interval. However, the heartbeat mechanism-based technique for fault detection suffers from weaknesses of either fast detection with low accuracy or completeness in detecting failures with a lengthy timeout. In this paper, we propose Affirmative Adaptive Failure Detection (AAFD). In this technique, the integration of newly proposed failure detection algorithm and the ping service is essential not only for dynamically improving certainty level of accuracy, but it is also very significant in verifying the aliveness of a site for strong completeness failure detection and reduces waiting time. The model outperforms the existing techniques by 18% to 39% in term of algorithm performance. On the average, AAFD detection is about 30% better than other detection algorithms.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: distributed system environment; fault detection; heartbeat methodology; system failure
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800-8360 Electronics
Divisions: Faculty of Computer Science and Information Technology > Department of Software Engineering
Depositing User: Normajihan Abd. Rahman
Date Deposited: 23 Apr 2015 00:36
Last Modified: 23 Apr 2015 00:36
URI: http://eprints.uthm.edu.my/id/eprint/3578
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item