Fault-Tolerance Techniques for High-Performance Computing

This timely text/reference presents a comprehensive overview of fault tolerance techniques for high-performance computing (HPC).

The text opens with a detailed introduction to the concepts of checkpoint protocols and scheduling algorithms, prediction, replication, silent error detection and correction, together with some application-specific techniques such as algorithm-based fault tolerance. Emphasis is placed on analytical performance models. This is then followed by a review of general-purpose techniques, including several checkpoint and rollback recovery protocols. Relevant execution scenarios are also evaluated and compared through quantitative models.

Topics and features: includes self-contained contributions from an international selection of preeminent experts; provides a survey of resilience methods and performance models; examines the various sources for errors and faults in large-scale systems, detailing their characteristics, with a focus on modeling, detection and prediction; reviews the spectrum of techniques that can be applied to design a fault-tolerant message passing interface; investigates different approaches to replication, comparing these to the traditional checkpoint-recovery approach; discusses the challenge of energy consumption of fault-tolerance methods in extreme-scale systems, proposing a methodology to estimate such energy consumption.

This authoritative volume is essential reading for all researchers and graduate students involved in high-performance computing.