A Control-Theoretic Design And Analysis Framework For Resilient Hard Real-Time Systems

<p>We introduce a new design metric called system-resiliency which characterizes the maximum unpredictable</p> <p>external stresses that any hard-real-time performance mode can withstand. Our proposed systemresiliency</p> <p>framework addresses resiliency determination for real-time systems with physical and hardware</p> <p>limitations. Furthermore, our framework advises the system designer about the feasible trade-offs between</p> <p>external system resources for the system operating modes on a real-time system that operates in a</p> <p>multi-parametric resiliency environment.</p> <p>Modern multi-modal real-time systems degrade the system’s operational modes as a response to unpredictable</p> <p>external stimuli. During these mode transitions, real-time systems should demonstrate a reliable</p> <p>and graceful degradation of service. Many control-theoretic-based system design approaches exist. Although</p> <p>they permit real-time systems to operate under various physical constraints, none of them allows</p> <p>the system designer to predict the system-resiliency over multi-constrained operating environment. Our</p> <p>framework fills this gap; the proposed framework consists of two components: the design-phase and runtime</p> <p>control. With the design-phase analysis, the designer predicts the behavior of the real-time system for</p> <p>variable external conditions. Also, the runtime controller navigates the system to the best desired target</p> <p>using advanced control-theoretic techniques. Further, our framework addresses the system resiliency of</p> <p>both uniprocessor and multicore processor systems.</p> <p>As a proof of concept, we first introduce a design metric called thermal-resiliency, which characterizes</p> <p>the maximum external thermal stress that any hard-real-time performance mode can withstand. We verify</p> <p>the thermal-resiliency for the external thermal stresses on a uniprocessor system through a physical testbed.</p> <p>We show how to solve some of the issues and challenges of designing predictable real-time systems that</p> <p>guarantee hard deadlines even under transitions between modes in an unpredictable thermal environment</p> <p>where environmental temperature may dynamically change using our new metric.</p> <p>We extend the derivation of thermal-resiliency to multicore systems and determine the limitations of</p> <p>external thermal stress that any hard-real-time performance mode can withstand. Our control-theoretic</p> <p>framework allows the system designer to allocate asymmetric processing resources upon a multicore proiii</p> <p>cessor and still maintain thermal constraints.</p> <p>In addition, we develop real-time-scheduling sub-components that are necessary to fully implement our</p> <p>framework; toward this goal, we investigate the potential utility of parallelization for meeting real-time</p> <p>constraints and minimizing energy. Under malleable gang scheduling of implicit-deadline sporadic tasks</p> <p>upon multiprocessors, we show the non-necessity of dynamic voltage/frequency regarding optimality of</p> <p>our scheduling problem. We adapt the canonical schedule for DVFS multiprocessor platforms and propose</p> <p>a polynomial-time optimal processor/frequency-selection algorithm.</p> <p>Finally, we verify the correctness of our framework through multiple measurable physical and hardware</p> <p>constraints and complete our work on developing a generalized framework.</p>

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Open Access Dissertation

Discipline thesis:degree_discipline

Computer Science

Year dc:date.available

2015

Author dc:creator

• Hettiarachchi, Pradeep Mahendra

Contributors dc:contributor

• Nathan Fisher

• Computer Sciences

Repository record dc:identifier

https://digitalcommons.wayne.edu/oa_dissertations/1339

OAI identifier oai:identifier

oai:digitalcommons.wayne.edu:oa_dissertations-2338