Comunicação Não Confiável em Detectores de Defeitos com Falhas por Crash

• Luiz Angelo Barchet Estefanel UFRGS
• Ingrid Jansch-Pôrto UFRGS

DOI: https://doi.org/10.5753/wtf.2000.23476

Abstract

The definition of failure detectors to be used on environments with crash failure hypothesis has some severe restrictions even in its weaker implementations. Yet, failure detectors are possible in distributed asynchronous systems, if we use technics to circumvent these restrictions, like the timeouts. During the construction of failure detectors, the need of reliable communication is an important factor: even if the definition if the failure detectors requires reliable communication, practical experiences show that we can use unreliable communication in its construction. This paper focuses on the native choices for reliable and unreliable communications, TCP and UDP, examining their impact on failure detectors' behavior.

References

AGUILERA, Marcos Kawazoe; CHEN, Wei; TOUEG, Sam. Heartbeat: a timeout-free failure detector for quiescent reliable communication. In: 11th International Workshop on Distributed Algorithms, Proceedings..., Setembro 1997. Também publicado como Technical Report, Comell University, Maio 1997. Disponível por WWW em [link] (11 Janeiro 2000)

CHANDRA, Tushar Deepak.; TOUEG, SAM. Unreliable Failure Detectors for Reliable Distributed Systems. Journal of the ACM, v. 43, n. 2, pg 225-267, Marco 1996. Também disponivel por WWW em [link] (12 Janeiro 2000)

DOUDOU, A.; GARBINATO, Benoit; GUERRAOUI, Rachid; and SCHIPER, André. Muteness Failure Detectors: Specification & Implementation. In: Third European Dependable Computing Conference (EDCC-3), Proceedings..., Springer Verlag. Suiça, Setembro 1999. Também disponível por WWW em [link] (9 Maio 2000)

FELBER, Pascal. The CORBA Object Group Service, Tese de Doutorado, École Polytechnique Fédérale de Lausanne, Suíça, 1998. Disponível por WWW em http://lsewww.epfl.ch/OGS/thesis/ (12 Janeiro 2000)

FISCHER, Michael J.; LYNCH, Nancy A.; PATERSON, Michael S. Impossibility of distributed consensus with one faulty process. Journal of the ACM, v. 32, n. 2, pg 374-382, 1985.

GARBINATO, Benoit. Protocol Objects and Patterns for Structuring Reliable Distributed Systems, Tese de Doutorado, Ecole Polytechnique Fédérale de Lausanne, Suiça, 1998. Disponivel por WWW em http://lsewww.epfl.ch/garbinato/PhD/ (11 Janeiro 2000).

GUERRAOUI, Rachid; SCHIPER, André. Consensus: The Big Misunderstanding. In: IEEE International Workshop on Future Trends in Distributed Computing Systems (FTDCS'97), Outubro 1997. Proceedings.... Também disponivel por WWW em http://lsewww.epfl.ch/~rachid/papers/ftdcs2-97.ps (12 Janeiro 2000)

MOSER, L. E.; MELLIAR-SMITH, P. M.; AGARWAL, D. A.; BUDHIA, R. K., LINGLEY-PAPADOPOULUS, C. A. Totem: A Fault-Tolerant Multicast Group Communication System. Communications of the ACM, v. 39, n. 4, pg 54-63, Abril 1996.

SCHIPER, André. Implementação de um detector Pull, Março 2000. Comunicação pessoal por e-mail.