Maximizing the Fault Tolerance Capability of Fixed Priority Schedules

Abstract

Component-Based Development (CBD) of software, with its successes in enterprise computing, has the promise of being a good development model due to its cost effectiveness and potential for achieving high quality of components by virtue of reuse. However, for systems with dependability concerns, such as real-time systems, a major challenge in using CBD consists of predicting dependability attributes, or providing dependability assertions, based on the individual component properties and architectural aspects. In this paper, we propose a framework which aims to address this challenge. Specifically, we present a revised error classification together with error propagation aspects, and briefly sketch how to compose error models within the context of Component-Based Systems (CBS). The ultimate goal is to perform the analysis on a given CBS, in order to find bottlenecks in achieving dependability requirements and to provide guidelines to the designer on the usage of appropriate error detection and fault tolerance mechanisms.

A Cascading Redundancy Approach for Dependable Real-Time Systems

“…In this paper, we propose a cascading redundancy framework capable of tolerating errors with a wider coverage (with respect to error frequency and error types) than time and space redundancy in isolation, handles tasks with mixed criticalities, and, above all, ensures that every critical task instance can be feasibly replicated in both time and space, independent of the RT scheduling policy. Hence, the research presented in this paper is significantly different from our previous works [2,8] due to the following original contributions:…”

“…An approach to maximize the fault tolerance capability in FPS has been presented in [8]. In this paper we propose a scheduler independent approach, suitable for, e.g., FPS, earliest deadline first (EDF) or table driven scheduling (TD), to enable time redundancy for every critical and ultra-critical task instance, executed along with highly-critical as well as non-critical tasks.…”

“…The new task's instances comprise all instances of the original tasks. A complete description of the method together with an example of the ILP problem formulation, as well as an evaluation of the approach in comparison with the fault tolerant version of rate monotonic algorithm, can be found in [8].…”

Task-Level Probabilistic Scheduling Guarantees for Dependable Real-Time Systems - A Designer Centric Approach

“…Lima and Burns [12], [13] extended this analysis in case of multiple faults, as well as for the case of increasing the priority of a critical task's alternate upon fault occurrences, and in [14] an upper bound for fault-tolerance in realtime systems based on slack redistribution is presented. We have proposed a work to maximize the fault tolerant capability of Fixed Priority Schedules [15], assuming each task instance is hit by an error and subsequently adapted this approach to Controller Area Network (CAN) in [3]. While the above works have advanced the field of fault tolerant scheduling within specified contexts, each one has some of the shortcomings such as lacking to provide probabilistic scheduling guarantees, restrictive task and fault models, non-consideration of multiple task criticality levels, high computational requirements of complex on-line mechanisms, and scheduler modifications which may be unacceptable from an industrial perspective.…”

Fault Tolerant Scheduling on Controller Area Network (CAN)

“…As a major concern is the number of priorities that may increase, we use ILP to find the priorities and the splits that yield the lowest number of messages that satisfy the inequalities, and implicitly the lowest number of priority levels. A description of the ILP problem formulation that can be easily adapted to CAN scheduling can be found in [17].…”