Journal of the ACM (JACM)
Diagnosis and reconfiguration in multiprocessor systems
Diagnosis and reconfiguration in multiprocessor systems
Optimal reconfiguration strategy for a degradable multimodule computing system
Journal of the ACM (JACM)
Surviving Errors in Component-Based Software
EUROMICRO '05 Proceedings of the 31st EUROMICRO Conference on Software Engineering and Advanced Applications
Diagnosis in the Presence of Known Faults
IEEE Transactions on Computers
Design patterns for graceful degradation
Transactions on Pattern Languages of Programming I
Hi-index | 0.00 |
Diagnosis and reconfiguration is concerned with the detection and location of faulty units and the minimization of the effect of faults within a digital system. This topic is motivated by the need for highly available systems that can continue essential operation, perhaps with degraded performance, when hardware failures occur. In recent years there has been a great progress in development of large-scale integrated (LSI) and microcomputer technology. This led to the consideration of systems of complex elements, such as networks of microprocessors. Today, multiprocessor structures are not only feasible but also highly attractive as cost-effective computer structures. Having a multiplicity of independent processors raises the opportunity for making the system self-diagnosable through interprocessor communication. Also the homogenity of the processors make it possible that after the occurrence of faults the system can reconfigure so that the effect of faults is minimized. In this paper we consider techniques for the design of systems which incorporate diagnostic and computational reconfiguration.