Title :
Recognition of catastrophic faults
Author :
Nayak, Amiya ; Pagli, Linda ; Santoro, Nicola
Author_Institution :
Center for Parallel & Distributed Comput., Sch. of Comput. Sci., Carleton Univ., Ottawa, Ont., Canada
Abstract :
For a given design, it is not difficult to identify a set of elements whose failure will have catastrophic consequence. There exist many patterns (random distribution) of faults, not in a block, which can be fatal for the system. Therefore, the characterization of such fault patterns is crucial for the identification, testing and detection of such catastrophic events. This paper, is concerned with the development of efficient recognition schemes; that is, efficient mechanisms which automatically determine whether or not an observed/detected pattern of faults will have catastrophic consequences. The problem of recognizing whether a fault pattern is catastrophic has been addressed
Keywords :
fault tolerant computing; reconfigurable architectures; system recovery; catastrophic faults; fault patterns; fault-tolerant computing; observed/detected pattern; random distribution; reconfiguration; Computer science; Distributed computing; Event detection; Fault detection; Fault diagnosis; Fault tolerance; Pattern recognition; Redundancy; Sufficient conditions; Testing;
Conference_Titel :
Defect and Fault Tolerance in VLSI Systems, 1992. Proceedings., 1992 IEEE International Workshop on
Conference_Location :
Dallas, TX
Print_ISBN :
0-8186-2837-5
DOI :
10.1109/DFTVS.1992.224370