Warns Structural Failure Models for Fault-Tolerant Distributed Computing

2010
ISBN: 978-3-8348-9707-7
Verlag: Vieweg & Teubner
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 216 Seiten, eBook

Reihe: Software Engineering Research

Timo Warns has developed tractable fault models that, while being non-probabilistic, are accurate for dependent and propagating faults. Using seminal problems such as consensus and constructing coteries, he demonstrates how the new models can be used to design and evaluate effective and efficient means of fault tolerance.

Dr. Timo Warns completed his doctoral thesis under the supervision of Prof. Dr. Wilhelm Hasselbring at the Carl von Ossietzky University of Oldenburg.

Zielgruppe

Research

Autoren/Hrsg.

Warns, Timo

Inhaltsverzeichnis

1;Foreword;6
2;Acknowledgments;7
3;Abstract;9
4;Zusammenfassung;11
5;Contents;13
6;List of Figures and Tables;15
7;1 Introduction;16
7.1;1.1 Motivation;16
7.2;1.2 Objectives;20
7.3;1.3 Outline;20
7.4;1.4 Remarks on Notation;23
8;2 Modelling Fault-Tolerant Distributed Systems;24
8.1;2.1 Interprocess Communication;25
8.2;2.2 States, Traces, Properties;27
8.3;2.3 Temporal Logic of Actions;34
8.4;2.4 Fault Model;40
8.5;2.5 Fault Tolerance;44
8.6;2.6 Timing Model;46
8.7;2.7 Summary;52
9;3 Modelling Fault Assumptions with Structural Failure Models;54
9.1;3.1 Related Work;55
9.2;3.2 Functional Failure Models;57
9.3;3.3 Structural Failure Models;62
9.4;3.4 Component Failure Models;65
9.4.1;3.4.1 Process Failure Models;66
9.4.2;3.4.2 Channel Failure Models;70
9.4.3;3.4.3 Hybrid Failure Models;73
9.5;3.5 Set-Based Structural Failure Models;75
9.5.1;3.5.1 Dependent Faults;76
9.5.2;3.5.2 Set-Based Structural Failure Models;78
9.5.3;3.5.3 Threshold Assumptions and Dependent Faults;87
9.5.4;3.5.4 Classes of Set-Based Structural Failure Models;89
9.6;3.6 Sequence-Based Structural Failure Models;104
9.7;3.7 Stochastics, Sets, and Sequences;107
9.8;3.8 Summary;109
10;4 Constructing Coteries;111
10.1;4.1 Related Work;113
10.2;4.2 Introduction to Quorums;115
10.3;4.3 Highly Available Static Coteries;123
10.4;4.4 Highly Available Dynamic Coteries;132
10.5;4.5 Reducing Probe Complexity;140
10.6;4.6 Summary;151
11;5 Reaching Consensus;153
11.1;5.1 Related Work;155
11.2;5.2 Introduction to Consensus;157
11.2.1;5.2.1 Specifying Consensus;157
11.2.2;5.2.2 The FLP Impossibility Result;161
11.2.3;5.2.3 Principles of Solving Consensus;162
11.2.4;5.2.4 Quality of Consensus Algorithms;164
11.3;5.3 Consensus in Asynchronous Systems with Unreliable Failure Detectors;168
11.3.1;5.3.1 A Versatile Consensus Algorithm;169
11.3.2;5.3.2 Adversary Structures;173
11.3.3;5.3.3 Didep Models;176
11.4;5.4 Consensus in Partially Synchronous Systems;183
11.5;5.5 Consensus in Synchronous Systems;188
11.6;5.6 Summary;200
12;6 Conclusion and Future Work;202
13;Bibliography;207
14;Index;224

Modelling Fault-Tolerant Distributed Systems.- Modelling Fault Assumptions with Structural Failure Models.- Constructing Coteries.- Reaching Consensus.- Conclusion and FutureWork.

Über Autor(innen)

Dr. Timo Warns completed his doctoral thesis under the supervision of Prof. Dr. Wilhelm Hasselbring at the Carl von Ossietzky University of Oldenburg.