Estimating Error-Probability and Its Application for Optimizing Roll-back Recovery with Checkpointing

The probability for errors to occur in electronic systems is not known in advance, but depends on many factors including influence from the environment where the system operates. In this paper, it is demonstrated that inaccurate estimates of the error probability lead to loss of performance in a well known fault tolerance technique, Roll-back Recovery with checkpointing (RRC). To regain the lost performance, a method for estimating the error probability along with an adjustment technique are proposed. Using a simulator tool that has been developed to enable experimentation, the proposed method is evaluated and the results show that the proposed method provides useful estimates of the error probability leading to near-optimal performance of the RRC fault-tolerant technique.©2010 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE. Dimitar Nikolov, Urban Ingelsson, Virendra Singh and Erik Larsson, Estimating Error-Probability and Its Application for Optimizing Roll-back Recovery with Checkpointing, 2010, 5th IEEE Intl. Symposium on Electronic Design, Test & Applications (DELTA 2010), Ho Chi Minh City, Vietnam, January 13-15, 2010, 281-285.