Automatically Increasing Fault Tolerance in Distributed Systems Automatically Increasing Fault Tolerance in Distributed Systems 2 System Model 7 3 Translations for Synchronous Systems: Deenitions and Limitations 17 6 Translations for Partially Synchronous Systems 81

Date Approved by Chairman Acknowledgments The guidance and encouragement of my advisor Gil Neiger were invaluable. For that and for his understanding, I thank him. Also, I would like to thank the members of my committee Mustaque Ahamad, Jim Burns, Ken Calvert, and H. Venkateswaran for their help and feedback. Many fellow students made my stay at Georgia Tech more enjoyable. Especially, I would like to thank I would like to thank the Hariri Foundation for making graduate school possible. In particular, I would like to thank Marc Muething and David Thompson at the Hariri Foundation for their help. Most of all, I would like to thank my wife Lina for her love, support and encouragement, not to mention her proofreading parts of the thesis. Last, but not least, I thank my parents for everything they have done for me. iii Contents Acknowledgments iii Summary vii 1 Introduction 1 1. Bibliography 90 Vita 93 v Summary Developing fault-tolerant distributed protocols is a diicult task. The diiculty of this task increases with the severity of the failures to be tolerated. One way to deal with this diiculty is to develop protocols tolerant of benign failures and then transform these protocols into ones that are tolerant of more severe failures. This transformation mechanism is called a translation. This dissertation considers a variety of processor failures and synchrony models. The failures studied range from simple stopping failures to arbitrary faulty behavior. The syn-chrony models range from systems in which processors are fully synchronized (synchronous systems) to systems in which processors are not synchronized at all (asynchronous systems). For all synchrony models, the dissertation gives general deenitions of translations and of measures to evaluate their performance. The two measures considered are communication complexity and fault-tolerance. Communication complexity is the communication overhead incurred when using a translation. Fault-tolerance is the maximum proportion of processors that can be faulty without aaecting the correctness of the translations. For synchronous systems, this dissertation presents a complete study of the relationship between fault-tolerance and round complexity of translations. It develops new translations that are optimal and proves that some previously developed translations are optimal. For asynchronous systems, it proves that some previously developed translations are optimal. For systems that are only partially synchronous this dissertation discusses some of the issues involved in designing eecient translations.