Failure detectors as first class objects

Felber, Pascal; Défago, Xavier; Guerraoui, Rachid; Oser, Pascal

doi:10.1109/doa.1999.794001

Cited by 74 publications

(42 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A hierarchy as proposed in [3] is not further investigated in this work, but could be easily built upon the monitoring groups which are introduced later on. Another hierarchical failure detector is presented by Felber et al [7]. They emphasise the importance of well defined interfaces for failure detectors being able to e.g.…”

Section: Related Workmentioning

confidence: 99%

Grouping Algorithms for Scalable Self-Monitoring Distributed Systems

Satzger¹,

Ungerer²

2008

Proceedings of the 2nd International ICST Conference on Autonomic Computing and Communication Systems

View full text Add to dashboard Cite

The growing complexity of distributed systems demands for new ways of control. Future systems should be able to adapt dynamically to the current conditions of their environment. They should be characterised by so-called self-x properties like self-configuring, self-healing, self-optimising, selfprotecting, and context-aware. For the incorporation of such features typically monitoring components provide the necessary information about the system's state. In this paper we propose three algorithms which allow a distributed system to install monitoring relations among its components. This serves as a basis to build scalable distributed systems with self-x features and to achieve a self-monitoring capability. Evaluation measurements have been conducted to compare the proposed algorithms.

show abstract

Section: Related Workmentioning

confidence: 99%

Grouping Algorithms for Scalable Self-Monitoring Distributed Systems

Satzger¹,

Ungerer²

2008

Proceedings of the 2nd International ICST Conference on Autonomic Computing and Communication Systems

View full text Add to dashboard Cite

show abstract

“…We used two implementations of failure detectors, one based on a push-style (FD-Push) and another based on a pull-style (FD-Pull) [5]. Both implementations are nonadaptive but use dynamic timeouts.…”

Section: Measuring the Impact Of Unreliable Failure Detectionmentioning

confidence: 99%

“…mes-sage per second, causing the predictor to be updated with the same frequency. Further, the predictor uses a safety margin based on a confidence interval of the round trip delay estimator (cib 5 ) [16]. CT-consensus and ACT-consensus were also simulated using AFD-Pull.…”

Section: An Architecture For Efficient and Robust Adaptive Consensus mentioning

confidence: 99%

Efficient and robust adaptive consensus services based on oracles

Sampaio¹,

Brasileiro²,

Nunes³

et al. 2005

J. Braz. Comp. Soc.

View full text Add to dashboard Cite

show abstract

“…The problem of failure detection has received much attention in the literature and many protocols mostly based on timeouts, have been proposed. The two classic models of fault detectors, discussed in [23], are the push and pull models. However, if these models have proved to work well in small networks, better proposals have been made for large-scale systems, such as the gossip-style protocol [24], which avoids the combinatorial explosion of messages.…”

Section: Fault Detection and Recoverymentioning

confidence: 99%