Scheduling algorithms for fault-tolerance in hard-real-time systems:

Abstract: "Many time-critical applications require high performance even if the system can suffer failures. This paper considers a distributed system with periodic tasks which can checkpoint their state on some reliable medium in order to handle failures. The problem of preemptive scheduling a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Bertossi, Alan A. (VerfasserIn), Mancini, Luigi V. (VerfasserIn)
Format: Buch
Sprache:English
Veröffentlicht: Pisa 1991
Schriftenreihe:Università degli Studi <Pisa> / Dipartimento di Informatica: Technical report 1991,15
Schlagworte:
Zusammenfassung:Abstract: "Many time-critical applications require high performance even if the system can suffer failures. This paper considers a distributed system with periodic tasks which can checkpoint their state on some reliable medium in order to handle failures. The problem of preemptive scheduling a set of such tasks is discussed where every occurrence of a task has to be completely executed before the next occurrence of the same task. A new class of scheduling algorithms is proposed which is based on a uniform rate strategy. This strategy yields optimal schedules in the absence of failures, and sub-optimal schedules when failures do occur. The performance of the solutions proposed is evaluated in terms of the number of processors and the number of preemptions needed
Moreover, analytical studies are used to reveal interesting trade-offs associated with the scheduling algorithms.
Beschreibung:20, [6] S. graph. Darst.

Es ist kein Print-Exemplar vorhanden.

Fernleihe Bestellen Achtung: Nicht im THWS-Bestand!