Wong, Tsz-Yeung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2002. / Includes bibliographical references (leaves 77-79). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Related Work --- p.2 / Chapter 1.2 --- Progressive Fault-Tolerant Mechanism --- p.4 / Chapter 1.3 --- Organization of This Thesis --- p.7 / Chapter 1.4 --- Contribution of The Thesis --- p.8 / Chapter 2 --- Server Failure Detection and Recovery --- p.9 / Chapter 3 --- Agent Failure Detection and Recovery --- p.12 / Chapter 3.1 --- System Architecture --- p.12 / Chapter 3.2 --- Protocol Design --- p.14 / Chapter 3.3 --- Failure and Recovery Scenarios --- p.16 / Chapter 3.3.1 --- When fails to receive msgiarrive --- p.17 / Chapter 3.3.2 --- Whenwi-1 fails to receive msgieave --- p.19 / Chapter 3.3.3 --- Failures of the witness agents and recovery scenarios --- p.22 / Chapter 3.3.4 --- Catastrophic failures --- p.24 / Chapter 3.4 --- Simplification --- p.24 / Chapter 4 --- Fault-Tolerant Mechanism Analysis --- p.27 / Chapter 4.1 --- Definitions and Notations --- p.27 / Chapter 4.2 --- Assumptions --- p.29 / Chapter 4.3 --- The Algorithm --- p.30 / Chapter 4.3.1 --- Informal algorithm descriptions --- p.30 / Chapter 4.3.2 --- Formal algorithm descriptions --- p.32 / Chapter 4.4 --- Liveness Proof --- p.39 / Chapter 4.5 --- Simplification Analysis --- p.52 / Chapter 5 --- Link Failure Analysis --- p.61 / Chapter 5.1 --- Problems of Link Failure --- p.61 / Chapter 5.2 --- Solution --- p.62 / Chapter 6 --- Reliability Evaluation --- p.67 / Chapter 6.1 --- Server Failure Detection Analysis --- p.68 / Chapter 6.2 --- Agent Failure Detection Analysis --- p.71 / Bibliography --- p.77 / A Glossary --- p.80

Typescript (photocopy). / Digitized by Kansas Correctional Industries

The development of fault-tolerant computing systems is a very difficult task. Two reasons contributed to this difficulty can be described as follows. The First is that, in normal practice, fault-tolerant computing policies and mechanisms are deeply embedded into most application programs, so that these application programs cannot cope with changes in environments, policies and mechanisms. These factors may change frequently in a distributed environment, especially in a heterogeneous environment. Therefore, in order to develop better fault-tolerant systems that can cope with constant changes in environments and user requirements, it is essential to separate the fault tolerant computing policies and mechanisms in application programs. The second is, on the other hand, a number of techniques have been proposed for the construction of reliable and fault-tolerant computing systems. Many computer systems are being developed to tolerant various hardware and software failures. However, most of these systems are to be used in specific application areas, since it is extremely difficult to develop systems that can be used in general-purpose fault-tolerant computing. The motivation of this thesis is based on these two aspects. The focus of the thesis is on developing a model based on the reactive system concepts for building better fault-tolerant computing applications. The reactive system concepts are an attractive paradigm for system design, development and maintenance because it separates policies from mechanisms. The stress of the model is to provide flexible system architecture for the general-purpose fault-tolerant application development, and the model can be applied in many specific applications. With this reactive system model, we can separate fault-tolerant computing polices and mechanisms in the applications, so that the development and maintenance of fault-tolerant computing systems can be made easier.

Graduation date: 1994

For a very large network deployed in space with only nearby nodes able to talk to each other, we want to do tasks like robust routing and data storage. One way to organize the network is via a hierarchy, but hierarchies often have a few critical nodes whose death can disrupt organization over long distances. I address this with a system of distributed aggregates called Persistent Nodes, such that spatially local failures disrupt the hierarchy in an area proportional to the diameter of the failure. I describe and analyze this system, which has been implemented in simulation.

Magnetic bearings are increasingly being used in a wide variety of applications in the industry such as compressors, turbines, motors, generators etc. Also, there are different types available depending upon their construction. The research presented here investigates a high temperature permanent magnet biased magnetic bearing system which is jointly being researched by National Aeronautics and Space Administration (NASA) and Electron Energy Corporation (EEC). The purpose of this research was to develop a permanent magnet biased magnetic bearing system using high temperature (HT) permanent magnets (PM) developed by EEC. This system was designed for high performance, high temperature (1000F) and high speed applications. The entire system consisted of two radial bearings, two catcher bearings, one axial thrust bearing and a motor. The central rotor shaft is powered by a high temperature permanent magnet motor to be able to run at the designed conditions of 20,000 rpm. This thesis documents the design of a feedback control law that stabilizes this HTPM biased AMB levitated system and summarizes efforts to build a test rig for the HT tests of the machine. A decentralized PD control law is used to achieve successful levitation. An existing PD analog controller with single input single output (SISO) control law for each axis (previously used for a flywheel test rig) is used as a feedback controller for this HTPM magnetic bearing system. Modeling and simulation of the resulting closed loop system is done in Matlab to test for stability and an iterative approach leads to optimum values of proportional and derivative gain pairs. The notch filter locations are also determined through this closed loop iterative simulation.

de Bruijn graphs are attractive due to its simplicity of routing messages between two nodes and the capability of fault tolerance. The shortest path from a node V to a node W in the directed binary de Bruijn graph can be obtained by firstly determining the longest substring, common to the right/left of V and to the left/right of W. Then L-operations/R-operations are performed to finish this routing process. However, this method does not always find the shortest path in the undirected binary de Bruijn graph. In this dissertation, we propose a shortest path routing algorithm which requires O(m2) time. We also design a fault-tolerant routing algorithm which provides the shortest path and another node-disjoint path of length at most m + log2m + 4. Our algorithm can tolerate one node failure in the m-dimensional binary de Bruijn network. In concurrent systems, a 1-fair alternator design is optimal if each processor can execute the critical step once in the fewest steps. This problem corresponds to use the minimum number of colors to color the processors in the system. Thus, the optimal design of a 1-fair alternator problem can be transformed into the coloring problem. We propose a simple and fast algorithm to solve the node coloring problem on the undirected binary de Bruijn graph. In our algorithm, the number of colors used is 3, and it is an optimal design. We also extend our method to solve the coloring problem on k-ary de Bruijn graphs. We first present a simple algorithm which needs 2k colors. By slight improvement, the number of required colors is reduced to k+1.

When the number of nodes increases, the chance that nodes or links fail increases. Then a fault-tolerant routing method is important to maintian the performance of the system. In the hypercube, safety levels and safety vectors provide the fault distribution information used to guide routing fault-tolerantly. The safety vectors for the hypercube describes the fault distribution more percisely than the safety level. The concept of safety levels has been applied to the star graph by other researchers. In this thesis, we apply the concept of the safety vectors in the hypercube to the star graph, and define three different safety vectors, including undirected safety vector, directed safety vector, and statistical safety vector. We first show the ability of the undirected safety vector. Then we extend the ideal to the directed safety vector and show it is better in deciding routing paths than the safety level for the star graph. We also show the reason that makes the directed safety vector not able to be used for derouting. In the previous result, a little change can make the directed safety vector usable for derouting in the hypercube. However, for the star graph, we can use only the information of neighbors to perform derouting with a slight modification in the directed safety vector. Then we set levels to the routing ability using the statistical safety vector. Try to make it contain more information of the fault distribution.

Thesis (M.S.)--Rutgers University, 2009. / "Graduate Program in Electrical and Computer Engineering." Includes bibliographical references (p. 60-62).

Thesis (Ph. D.)--University of Hong Kong, 2009. / Includes bibliographical references (leaves 92-100). Also available in print.