Fault-Tolerant Rate-Monotonic Scheduling

  • Authors:
  • Sunondo Ghosh;Rami Melhem;Daniel Mossé;Joydeep Sen Sarma

  • Affiliations:
  • Department of Computer Science, University of Pittsburgh, Pittsburgh, PA 15260;Department of Computer Science, University of Pittsburgh, Pittsburgh, PA 15260;Department of Computer Science, University of Pittsburgh, Pittsburgh, PA 15260;Department of Computer Science, University of Pittsburgh, Pittsburgh, PA 15260

  • Venue:
  • Real-Time Systems
  • Year:
  • 1998

Quantified Score

Hi-index 0.01

Visualization

Abstract

Due to the critical nature of the tasks in hard real-timesystems, it is essential that faults be tolerated. In this paper,we present a scheme which can be used to tolerate faults duringthe execution of preemptive real-time tasks. We describe a recoveryscheme which can be used to re-execute tasks in the event ofsingle and multiple transient faults and discuss conditions thatmust be met by any such recovery scheme. We then extend the originalRate Monotonic Scheduling (RMS) scheme and the exact characterizationof RMS to provide tolerance for single and multiple transientfaults. We derive schedulability bounds for sets of real-timetasks given the desired level of fault tolerance for each taskor subset of tasks. Finally, we analyze and compare those boundswith existing bounds for non-fault-tolerant and other variationsof RMS.