Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Real-Time Syst (2011) 47: 109142 DOI 10.1007/s11241-011-9117-x

Summary: Real-Time Syst (2011) 47: 109142
DOI 10.1007/s11241-011-9117-x
Global scheduling based reliability-aware power
management for multiprocessor real-time systems
Xuan Qi Dakai Zhu Hakan Aydin
Published online: 5 February 2011
Springer Science+Business Media, LLC 2011
Abstract Reliability-aware power management (RAPM) has been a recent research
focus due to the negative effects of the popular power management technique dy-
namic voltage and frequency scaling (DVFS) on system reliability. As a result, sev-
eral RAPM schemes have been studied for uniprocessor real-time systems. In this
paper, for a set of frame-based independent real-time tasks running on multiproces-
sor systems, we study global scheduling based RAPM (G-RAPM) schemes. Depend-
ing on how recovery blocks are scheduled and utilized, both individual-recovery and
shared-recovery based G-RAPM schemes are investigated. An important dimension
of the G-RAPM problem is how to select the appropriate subset of tasks for en-
ergy and reliability management (i.e., scale down their executions while ensuring
that they can be recovered from transient faults). We show that making such decision
optimally (i.e., the static G-RAPM problem) is NP-hard. Then, for the individual-
recovery based approach, we study two efficient heuristics, which rely on local and


Source: Aydin, Hakan - Department of Computer Science, George Mason University


Collections: Computer Technologies and Information Sciences; Engineering