Figures xv
 
Tables xix
 
Equations xxi
 
Preface and Acknowledgments xxiii
 
Audience xxiv
 
Organization xxiv
 
Acknowledgments xxvi
 
PART 1 BASICS 1
 
1 SERVICE, RISK, AND BUSINESS CONTINUITY 3
 
1.1 Service Criticality and Availability Expectations 3
 
1.2 The Eight-Ingredient Model 4
 
1.3 Catastrophic Failures and Geographic Redundancy 7
 
1.4 Geographically Separated Recovery Site 11
 
1.5 Managing Risk 12
 
1.6 Business Continuity Planning 14
 
1.7 Disaster Recovery Planning 15
 
1.8 Human Factors 17
 
1.9 Recovery Objectives 17
 
1.10 Disaster Recovery Strategies 18
 
2 SERVICE AVAILABILITY AND SERVICE RELIABILITY 20
 
2.1 Availability and Reliability 20
 
2.2 Measuring Service Availability 25
 
2.3 Measuring Service Reliability 33
 
PART 2 MODELING AND ANALYSIS OF REDUNDANCY 35
 
3 UNDERSTANDING REDUNDANCY 37
 
3.1 Types of Redundancy 37
 
3.2 Modeling Availability of Internal Redundancy 44
 
3.3 Evaluating High-Availability Mechanisms 52
 
4 OVERVIEW OF EXTERNAL REDUNDANCY 59
 
4.1 Generic External Redundancy Model 59
 
4.2 Technical Distinctions between Georedundancy and Co-Located Redundancy 74
 
4.3 Manual Graceful Switchover and Switchback 75
 
5 EXTERNAL REDUNDANCY STRATEGY OPTIONS 77
 
5.1 Redundancy Strategies 77
 
5.2 Data Recovery Strategies 79
 
5.3 External Recovery Strategies 80
 
5.4 Manually Controlled Recovery 81
 
5.5 System-Driven Recovery 83
 
5.6 Client-Initiated Recovery 85
 
6 MODELING SERVICE AVAILABILITY WITH EXTERNAL SYSTEM REDUNDANCY 98
 
6.1 The Simplistic Answer 98
 
6.2 Framing Service Availability of Standalone Systems 99
 
6.3 Generic Markov Availability Model of Georedundant Recovery 103
 
6.4 Solving the Generic Georedundancy Model 115
 
6.5 Practical Modeling of Georedundancy 121
 
6.6 Estimating Availability Benefit for Planned Activities 130
 
6.7 Estimating Availability Benefit for Disasters 131
 
7 UNDERSTANDING RECOVERY TIMING PARAMETERS 133
 
7.1 Detecting Implicit Failures 134
 
7.2 Understanding and Optimizing RTO 141
 
8 CASE STUDY OF CLIENT-INITIATED RECOVERY 147
 
8.1 Overview of DNS 147
 
8.2 Mapping DNS onto Practical Client-Initiated Recovery Model 148
 
8.3 Estimating Input Parameters 154
 
8.4 Predicted Results 165
 
8.5 Discussion of Predicted Results 172
 
9 SOLUTION AND CLUSTER RECOVERY 174
 
9.1 Understanding Solutions 174
 
9.2 Estimating Solution Availability 177
 
9.3 Cluster versus Element Recovery 179
 
9.4 Element Failure and Cluster Recovery Case Study 182
 
9.5 Comparing Element and Cluster Recovery 186
 
9.6 Modeling Cluster Recovery 187
 
PART 3 RECOMMENDATIONS 201
 
10 GEOREDUNDANCY STRATEGY 203
 
10.1 Why Support Multiple Sites? 203
 
10.2 Recovery Realms 204
 
10.3 Recovery Strategies 206
 
10.4 Limp-Along Architectures 207
 
10.5 Site Redundancy Options 208
 
10.6 Virtualization, Cloud Computing, and Standby Sites 216
 
10.7 Recommended Design Methodology 217
 
11 MAXIMIZING SERVICE AVAILABILITY VIA GEOREDUNDANCY 219
 
11.1 Theoretically Optimal External Redundancy 219
 
11.2 Practically Optimal Recovery Strategies 220
 
11.3 Other Considerations 228
 
12 GEOREDUNDANCY REQUIREMENTS 230
 
12.1 Internal Redundancy Requir