By Topic

Optimizing resources in real-time scheduling for fault tolerant processors

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Pillay, R. ; Amrita Vishwa Vidyapeetham, Coimbatore, India ; Chandran, S.K. ; Punnekkat, S.

The safety critical systems used in avionics, nuclear power plants and emergency medical equipments have to meet stringent reliability and temporal demands. Such demands are met with fault tolerant mechanisms, such as hardware and software redundancy. In this paper, we consider a safety critical application, the dual redundant onboard computer (OBC) system of the Indian Satellite Launch Vehicle and propose a scheme to optimize the onboard computing resources without detracting from the system reliability requirements. The redundancy is dealt with at the task allocation level and the slack generated, is used for allocation of more computational tasks, making the scheme very attractive in terms of efficient management of resources. The scheme of task allocation combined with real-time scheduling using Rate Monotonic (RM) and Earliest Deadline First (EDF) provide more programming flexibility and efficiently utilize the system resources. The scheme when implemented gives an efficient offline task allocation for fault-free conditions and flexible fault tolerance strategy during processor failure. The proposed scheme is compared with a traditional dual scheme. The implementation is experimented with a simulation and evaluated using performance metrics to illustrate the enhanced performance capability of the approach. This scheme, extended to multiprocessors with generic features can lead to tremendous throughput in terms of performance and costs. The contributions of this work are a system level algorithm for the implementation of real-time task allocation and scheduling.

Published in:

Parallel Distributed and Grid Computing (PDGC), 2010 1st International Conference on

Date of Conference:

28-30 Oct. 2010