Reliability-Aware Proactive Energy Management in Hard Real-Time Systems: A Motivational Case Study

Reliability-Aware Proactive Energy Management in Hard Real-Time Systems: A Motivational Case Study

Satyakiran Munaga, Francky Catthoor
Copyright: © 2010 |Volume: 1 |Issue: 4 |Pages: 11
ISSN: 1947-9220|EISSN: 1947-9239|EISBN13: 9781613502723|DOI: 10.4018/jaras.2010100101
Cite Article Cite Article

MLA

Munaga, Satyakiran, and Francky Catthoor. "Reliability-Aware Proactive Energy Management in Hard Real-Time Systems: A Motivational Case Study." IJARAS vol.1, no.4 2010: pp.1-11. http://doi.org/10.4018/jaras.2010100101

APA

Munaga, S. & Catthoor, F. (2010). Reliability-Aware Proactive Energy Management in Hard Real-Time Systems: A Motivational Case Study. International Journal of Adaptive, Resilient and Autonomic Systems (IJARAS), 1(4), 1-11. http://doi.org/10.4018/jaras.2010100101

Chicago

Munaga, Satyakiran, and Francky Catthoor. "Reliability-Aware Proactive Energy Management in Hard Real-Time Systems: A Motivational Case Study," International Journal of Adaptive, Resilient and Autonomic Systems (IJARAS) 1, no.4: 1-11. http://doi.org/10.4018/jaras.2010100101

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Advanced technologies such as sub-45nm CMOS and 3D integration are known to have more accelerated and increased number of reliability failure mechanisms. Classical reliability assessment methodology, which assumes ad-hoc failure criteria and worst-case for all influencing dynamic aspects, is no longer viable in these technologies. In this paper, the authors advocate that managing temperature and reliability at run-time is necessary to overcome this reliability-wall without incurring significant cost penalty. Nonlinear nature of modern systems, however, makes the run-time control very challenging. The authors suggest that full cost-consciousness requires a truly proactive controller that can efficiently manage system slack with future in perspective. This paper introduces the concept of “gas-pedal,” which enhances the effectiveness of the proactive controller in minimizing the cost without sacrificing the hard guarantees required by the constraints. Reliability-aware dynamic energy management of a processor running AVC motion compensation task is used as a motivational case study to illustrate the proposed concepts.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.