Title :
Performability-driven adaptive fault tolerance
Author_Institution :
Dept. of Comput. Sci., Texas Univ., Dallas, TX, USA
Abstract :
We introduce a performability-driven realization of adaptive fault tolerance. The object system for the investigation is a multiprocessor system in a real-time context accommodating parallel computation and two fault tolerance schemes, namely, recovery blocks (RB) and distributed recovery blocks (DRB). The optimality criterion is directly specified in terms of a preferability measure, based on which adaptive rules are generated to govern the utilization of RB and DRB during system´s operation. The analysis shows that performability concepts and modeling methods enable us to manage the tradeoffs among system attributes in a formal manner and lead to effective realization of adaptive fault tolerance.<>
Keywords :
fault tolerant computing; multiprocessing systems; reliability; distributed recovery blocks; multiprocessor system; object system; optimality criterion; parallel computation; performability-driven adaptive fault tolerance; real-time context; recovery blocks; system attributes; Computer science; Concurrent computing; Degradation; Environmental management; Fault tolerance; Fault tolerant systems; Multiprocessing systems; Performance analysis; Performance evaluation; Real time systems;
Conference_Titel :
Fault-Tolerant Computing, 1994. FTCS-24. Digest of Papers., Twenty-Fourth International Symposium on
Conference_Location :
Austin, TX, USA
Print_ISBN :
0-8186-5520-8
DOI :
10.1109/FTCS.1994.315643
