Fault-tolerant Hard Real-time Systems Research Articles

Two-State Checkpointing for Energy-Efficient Fault Tolerance in Hard Real-Time Systems

Checkpointing with rollback recovery is a well-established technique to tolerate transient faults. However, it incurs significant time and energy overheads, which go wasted in fault-free execution states and may not even be feasible in hard real-time systems. This paper presents a low-overhead two-state checkpointing (TsCp) scheme for fault-tolerant hard real-time systems. It differentiates between the fault-free and faulty execution states and leverages two types of checkpoint intervals for these two different states. The first type is nonuniform intervals that are used while no fault has occurred. These intervals are determined based on postponing checkpoint insertions in fault-free states, with the aim of decreasing the number of checkpoint insertions. The second type is uniform intervals that are used from the time when the first fault occurs. They are determined so as to minimize execution time for faulty states, leaving more time available for energy management in fault-free states. Experimental evaluation on an embedded processor (LEON3) and an emerging nonvolatile memory technology (ReRAM) illustrates that TsCp significantly reduces the number of checkpoints (62% on average) compared with previous works, while preserving fault tolerance. This results in 14% and 13% reduced execution time and energy consumption, respectively. Furthermore, we combine TsCp with dynamic voltage scaling (DVS) and achieve up to 26% (21% on average) energy saving compared with the state-of-the-art techniques.

Discrete feedback-based dynamic voltage scaling for safety critical real-time systems

Recently, the tradeoff between low energy consumption and high fault-tolerance has attracted a lot of attention as a key issue in the design of real-time systems. Dynamic Voltage Scaling (DVS) is commonly employed as one of the most effective low energy techniques for real-time systems. It has been observed that the use of feedback-based methods can improve the effectiveness of DVS-enabled systems. In this paper, we have investigated reducing the energy consumption of fault-tolerant hard real-time systems using the feedback control theory. Our proposed method makes the system capable of selecting the proper frequency and voltage settings in order to reduce the energy consumption, while guaranteeing hard real-time requirements in the presence of unpredictable workload fluctuations and faults. In the proposed method, the available slack-time is exploited by a feedback-based DVS at runtime to reduce the energy consumption. Furthermore, some slack-time is reserved for re-execution in case of faults. The simulation results show that compared with the traditional DVS method, our proposed method not only provides up to 59% energy saving, but also satisfies hard real-time constraints. Our proposed method is also effective in harnessing the static energy. The transition overheads are also taken into account in our simulation experiments.

Design Optimization of Time- and Cost-Constrained Fault-Tolerant Embedded Systems With Checkpointing and Replication

We present an approach to the synthesis of fault-tolerant hard real-time systems for safety-critical applications. We use checkpointing with rollback recovery and active replication for tolerating transient faults. Processes and communications are statically scheduled. Our synthesis approach decides the assignment of fault-tolerance policies to processes, the optimal placement of checkpoints and the mapping of processes to processors such that multiple transient faults are tolerated and the timing constraints of the application are satisfied. We present several design optimization approaches which are able to find fault-tolerant implementations given a limited amount of resources. The developed algorithms are evaluated using extensive experiments, including a real-life example.

An optimal fixed-priority assignment algorithm for supporting fault-tolerant hard real-time systems

The main contribution of this paper is twofold. First, we present an appropriate schedulability analysis, based on response time analysis, for supporting fault-tolerant hard real-time systems. We consider systems that make use of error-recovery techniques to carry out fault tolerance. Second, we propose a new priority assignment algorithm which can be used, together with the schedulability analysis, to improve system fault resilience. These achievements come from the observation that traditional priority assignment policies may no longer be appropriate when faults are being considered. The proposed schedulability analysis takes into account the fact that the recoveries of tasks may be executed at higher priority levels. This characteristic is very important since, after an error, a task certainly has a shorter period of time to meet its deadline. The proposed priority assignment algorithm, which uses some properties of the analysis, is very efficient. We show that the method used to find out an appropriate priority assignment reduces the search space from O(n!) to O(n/sup 2/), where n is the number of task recovery procedures. Also, we show that the priority assignment algorithm is optimal in the sense that the fault resilience of task sets is maximized as for the proposed analysis. The effectiveness of the proposed approach is evaluated by simulation.

Stabilizing Pre-Run-Time Schedules With the Help of GraceTime

This paper discusses the stability of a feasible pre-run-time schedule for a transient overload introduced by processes re-execution during an error recovery action. It shows that the stability of a schedule strictly tuned to meet hard deadlines is very small, invalidating thus backward error recovery. However, the stability of the schedule always increases when a real-time process is considered as having a nominal and a hard deadline separated by a non-zero grace time. This is true for sets of processes having arbitrary precedence and exclusion constraints, and executed on a single or multiprocessor based architecture. Grace time is not just the key element for the realistic estimation of the timing constraints of real-time error processing techniques. It also allows backward error recovery to be included in very efficient pre-run-time scheduled systems when the conditions stated in this paper are satisfied. This is a very important conclusion, as it shows that fault-tolerant hard real-time systems do not have to be extremely expensive and complex.