Preemptive Uniprocessor Scheduling of Mixed-Criticality Sporadic Task Systems

• Published in: Journal of the ACM Vol. 62; no. 2; pp. 1 - 33
• Main Authors: Baruah, Sanjoy, Bonifaci, Vincenzo, D'angelo, Gianlorenzo, Li, Haohan, Marchetti-Spaccamela, Alberto, Van Der Ster, Suzanne, Stougie, Leen
• Format: Journal Article
• Language: English
• Published: New York, NY, USA ACM 01.05.2015; Association for Computing Machinery
• Subjects: Algorithms; Approximation algorithms analysis; Certification; Computer Science; Computer systems organization; Contextual software domains; Data Structures and Algorithms; Design and analysis of algorithms; Embedded and cyber-physical systems; Information systems; Machine learning theory; Mathematical software; Mathematics of computing; Microprocessors; Online algorithms; Online learning algorithms; Operating systems; Optimization; Preempting; Process management; Real-time systems; Reinforcement learning; Scheduling; Scheduling algorithms; Sequential decision making; Software and its engineering; Software organization and properties; Studies; Task scheduling; Tasks; Theory and algorithms for application domains; Theory of computation; preemptive scheduling; sporadic task system; Mixed criticality
• ISSN: 0004-5411; 1557-735X; 1557-735X
• DOI: 10.1145/2699435

Systems in many safety-critical application domains are subject to certification requirements. For any given system, however, it may be the case that only a subset of its functionality is safety-critical and hence subject to certification; the rest of the functionality is non-safety-critical and does not need to be certified, or is certified to lower levels of assurance. The certification-cognizant runtime scheduling of such mixed-criticality systems is considered. An algorithm called EDF-VD (for Earliest Deadline First with Virtual Deadlines) is presented: this algorithm can schedule systems for which any number of criticality levels are defined. Efficient implementations of EDF-VD, as well as associated schedulability tests for determining whether a task system can be correctly scheduled using EDF-VD, are presented. For up to 13 criticality levels, analyses of EDF-VD, based on metrics such as processor speedup factor and utilization bounds, are derived, and conditions under which EDF-VD is optimal with respect to these metrics are identified. Finally, two extensions of EDF-VD are discussed that enhance its applicability. The extensions are aimed at scheduling a wider range of task sets, while preserving the favorable worst-case resource usage guarantees of the basic algorithm.