Scheduling and Optimization of Fault-Tolerant Embedded Systems with Transparency/Performance Trade-Offs

Abstract

In this article, we propose a strategy for the synthesis of fault-tolerant schedules and for the mapping of fault-tolerant applications. Our techniques handle transparency/performance trade-offs and use the faultoccurrence information to reduce the overhead due to fault tolerance. Processes and messages are statically scheduled, and we use process reexecution for recovering from multiple transient faults. We propose a finegrained transparent recovery, where the property of transparency can be selectively applied to processes and messages. Transparency hides the recovery actions in a selected part of the application so that they do not affect the schedule of other processes and messages. While leading to longer schedules, transparent recovery has the advantage of both improved debuggability and less memory needed to store the faulttolerant schedules. ACM Reference Format:Izosimov, V., Pop, P., Eles, P., and Peng, Z. 2012. Scheduling and optimization of fault-tolerant embedded systems with transparency/performance trade-offs.

An Energy-Efficient Fault-Tolerant Scheduling Algorithm Based on Variable Data Fragmentation

“…In [2], faults of buses are tolerated using a TDMA (Time Division Multiple Access) communication protocol and an active redundancy approach. In [10] authors propose a fine grained transparent recovery, where the property of transparency can be selectively applied to processes and messages. In [11] authors survey the problem of how to schedule tasks in such a way that deadlines continue to be met despite processor and communication media (permanent or transient) or software failure.…”

Value-based scheduling of distributed fault-tolerant real-time systems with soft and hard timing constraints

“…Such faults appear for a short time without causing permanent damage, and can be caused by electromagnetic interference, radiation, temperature variations, software "bugs", etc. [10]. Online scheduling with fault tolerance constraints for hard real-time systems has been considered in [8,16,23].…”

Design Optimization of Time-and Cost-Constrained Fault-Tolerant Distributed Embedded Systems

“…We have extended (2) to consider faults and detailed checkpointing overheads and for process , when process is considered in isolation (see [16] for the proof of the formula) if if (3) Equation (3) allows us to calculate the optimal number of checkpoints for a certain process considered in isolation. However, calculating the number of checkpoints for each individual process will not produce a solution which is globally optimal for the whole application because processes share recovery slacks.…”

“…In that case, a successor process of a replicated process can be placed in the root schedule at the earliest time moment , at which at least one valid message can arrive from a replica of process . 5 We also include in the set of valid messages the output from replica to successor passed through the shared memory (if replica and successor are mapped on the same computation node) [16].…”

“…The contingency schedules are determined online, in linear time, by the scheduler in each node, based on the occurrences of faults. The overhead due to obtaining the new start times is considered at the schedule table construction [16]. …”