Computing System Reliability: Models and Analysis (Cell Engineering)

Computing systems are of growing importance because of their wide use in many areas including those in safety-critical systems. This book describes the basic models and approaches to the reliability analysis of such systems. An extensive review is provided and models are categorized into different types. Some Markov models are extended to the analysis of some specific computing systems such as combined software and hardware, imperfect debugging processes, failure correlation, multi-state systems, heterogeneous subsystems, etc. One of the aims of the presentation is that based on the sound analysis and simplicity of the approaches, the use of Markov models can be better implemented in the computing system reliability. Team DDU......Page 1 Contents......Page 12 1.1. Need for Computing System Reliability Analysis......Page 16 1.2. Computing System Reliability Concepts......Page 17 1.3. Approaches to Computing System Modeling......Page 18 2.1. Reliability Measures......Page 22 2.2. Common Techniques in Reliability Analysis......Page 27 2.3. Markov Process Fundamentals......Page 34 2.4. Nonhomogeneous Poisson Process (NHPP) Models......Page 51 3.1. Single Component System......Page 56 3.2. Parallel Configurations......Page 63 3.3. Load-Sharing Configurations......Page 73 3.4. Standby Configurations......Page 76 3.5. Notes and References......Page 84 4.1. Basic Markov Model......Page 86 4.2. Extended Markov Models......Page 91 4.3. Modular Software Systems......Page 105 4.4. Models for Correlated Failures......Page 109 4.5. Software NHPP Models......Page 116 4.6. Notes and References......Page 125 5.1. Single-Processor System......Page 128 5.2. Models for Modular System......Page 137 5.3. Models for Clustered System......Page 143 5.4. A Unified NHPP Markov Model......Page 154 5.5. Notes and References......Page 158 6 AVAILABILITY AND RELIABILITY OF DISTRIBUTED COMPUTING SYSTEMS......Page 160 6.1. Introduction to Distributed Computing......Page 161 6.2. Distributed Program and System Reliability......Page 163 6.3. Homogeneously Distributed Software/Hardware Systems......Page 178 6.4. Centralized Heterogeneous Distributed Systems......Page 186 6.5. Notes and References......Page 191 7 RELIABILITY OF GRID COMPUTING SYSTEMS......Page 194 7.1. Introduction of the Grid Computing System......Page 195 7.2. Grid Reliability of the Resource Management System......Page 199 7.3. Grid Reliability of the Network......Page 203 7.4. Grid Reliability of the Software and Resources......Page 216 7.5. Notes and References......Page 219 8.1. Basic Concepts of Multi-State System (MSS)......Page 222 8.2. Basic Models for MSS Reliability......Page 229 8.3. A MSS Failure Correlation Model......Page 239 8.4. Notes and References......Page 251 9 OPTIMAL SYSTEM DESIGN AND RESOURCE ALLOCATION......Page 254 9.1. Optimal Number of Hosts......Page 255 9.2. Resource Allocation - Independent Modules......Page 262 9.3. Resource Allocation - Dependent Modules......Page 273 9.4. Optimal Design of the Grid Architecture......Page 281 9.5. Optimal Integration of the Grid Services......Page 284 9.6. Notes and References......Page 287 References......Page 290 Subject Index......Page 306