Low-Energy Standby-Sparing for Hard Real-Time Systems

Abstract

Abstract-Time-redundancy techniques are commonly used in real-time systems to achieve fault tolerance without incurring high energy overhead. However, reliability requirements of hard real-time systems that are used in safety-critical applications are so stringent that time-redundancy techniques are sometimes unable to achieve them. Standby sparing as a hardwareredundancy technique can be used to meet high reliability requirements of safety-critical applications. However, conventional standby-sparing techniques are not suitable for lowenergy hard real-time systems as they either impose considerable energy overheads or are not proper for hard timing constraints. In this paper we provide a technique to use standby sparing for hard real-time systems with limited energy budgets. The principal contribution of this work is an online energymanagement technique which is specifically developed for standby-sparing systems that are used in hard real-time applications. This technique operates at runtime and exploits dynamic slacks to reduce the energy consumption while guaranteeing hard deadlines. We compared the low-energy standby-sparing (LESS) system with a low-energy timeredundancy system (from a previous work). The results show that for relaxed time constraints, the LESS system is more reliable and provides about 26% energy saving as compared to the time-redundancy system. For tight deadlines when the timeredundancy system is not sufficiently reliable (for safety-critical application), the LESS system preserves its reliability but with about 49% more energy consumption.

Two-Phase Low-Energy N-Modular Redundancy for Hard Real-Time Multi-Core Systems

“…Some research works, e.g., [6], [7], [11], have addressed both fault tolerance and low energy-consumption in faulttolerant real-time systems with two processors. These works have not considered multiple faults per task execution, and also they assume they have perfect faultdetection mechanisms.…”

“…While such architectures have been employed for embedded applications that require high performance computing, we believe they also offer new considerable opportunities for designing embedded systems where hard real-time operation, high reliability in the presence of transient faults, and low energy consumption are required [6], [7], [8]. In this paper, we address the use of multi-core platforms to achieve high reliability with low energy-overhead for hard realtime embedded systems.…”

“…An NMR system can mask faults while less than half of its units are faulty. Fault-tolerant real-time systems that has been considered in previous works require faultdetection mechanisms (e.g., [5], [6], [7], [8], [11]) and these works have assumed (usually implicitly) that they have perfect detection mechanisms (i.e., they can detect all faulty task executions). However, common fault-detection mechanisms are far less effective than what is required for highly reliable systems, whereas NMR does not require any specific fault-detection mechanism and uses result comparison (majority voting) for fault-detection and masking [9], [10].…”

“…To reduce the energy overhead, we propose an energy-management technique that bears the major contributions of the work and is specifically developed for NMR when used for hard realtime multi-core systems (Sections 3 and 4). The main contributions of this work are: i) Considering the dominance of the fault-free execution on faulty executions [6] [8] [38], a two-phase NMR is proposed that achieves minimized energy consumption in the absence of faults while guaranteeing reliability and deadline requirements. ii) A specific type of slack time, called pseudo dynamic slack, is considered in this work.…”

“…ii) A specific type of slack time, called pseudo dynamic slack, is considered in this work. As explained in Section 4, this type of slack time is different from conventional slack times, i.e., static and dynamic slack [6], [8], [20].…”