Loading...

Reliablity-Aware Energy Management for Mixed-Criticality Systems on Multicores

Naghavi, Amin | 2020

545 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 52729 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Hesabi, Shahin
  7. Abstract:
  8. Integrating functionalities of different-criticality levels on a shared computing platform known as Mixed-Criticality Systems (MCSs) has been noticed recently in research and industrial designs. Due to the battery-operated nature of some MCSs and different reliability requirement for tasks, joint energy and reliability management is crucial in these systems. Another important issue in these systems which is rarely addressed in previous works is tolerating permanent faults. In this thesis, we propose two comprehensive schemes: MC-4S and MC-2S which guarantee to tolerate permanent faults and maintaining the system reliability with respect to the transient faults. In addition, guaranteeing the quality of service of low-criticality tasks in the case of permanent fault occurrence and improving the system’s total quality of service in case of overrun are other responsibilities of these schemes. In these schemes, we consider two copies for each high-criticality task on different cores to guarantee that all of them will meet their deadlines in case of permanent fault occurrence. However, in order to reduce the timing and power overheads, the alternative copies of low-criticality tasks are managed to execute only when a permanent fault or overrun occurs on their primary core. In MC-2S partitioned scheduling is employed and sufficient slack time is reserved for each low-criticality job on an alternative core, on the other hand, MC-4S uses semi-partitioned scheduling to manage low-criticality tasks of a core in which permanent fault or overrun occurs. In this paper, we also exploit a Demand Bound Function (DBF) based schedulability test to guarantee the timeliness, and use a preference-oriented scheduling algorithm with the proposed greedy reliability-aware DVFS method for more energy saving. We also examine the efficiency of our methods through energy consumption and acceptance ratio
  9. Keywords:
  10. Fault Tolerance ; Service Quality ; Reliability Aware Energy Management ; Multi-Core Platforms ; Multicore Embedded System ; Mixed-Criticality Emdedded Systems

 Digital Object List

 Bookmark

No TOC