Early comparison and decision strategies for datapaths that recover from transient faults

Abstract

Rollback and retry is a common approach used to achieve error recovery in datapaths that tolerate transient faults. In this approach, each segment of a computation is duplicated and the results are compared using fault-tolerant comparators. If the compared values are unequal, the segment is rolled back to the preceding correct state (rollback point) and retried from that state. We introduce early comparison and rollback strategies for use in such datapaths. These strategies utilize comparators during the computational portion of the segment and can initiate a rollback before the segment is completed. We illustrate through examples how these strategies can reduce hardware costs (number of comparators needed) and the delay in recovering from a transient fault compared to conventional strategies.Index Terms-Fault tolerance, high level synthesis, rollback and retry.

Performance Analysis and Validation of a Recoverable Flight Control System in a Simulated Neutron Environment

“…12,13 It has most recently been used to study recovery from transient faults introduced by high intensity electromagnetic radiation. 14,15 The error recovery technique implemented on the RCS is a variation of rollback recovery; 16,17 it has the following steps: checkpointing, faulttolerant comparison, rollback, and retry. During a checkpoint, the state of each microprocessor module is stored.…”

Stability analysis of upset recovery methods for electromagnetic interference [flight control systems]

“…During this period of time many studies of the performance of the recovery system have appeared in the literature (see eg. [18,19]), giving statistics of the recovery time or a procedure to minimize it. However, most of these studies are of limited use in real-time control applications since they do not take into account the effect of the recovery process in the closed-loop dynamical system.…”