Solving Enterprise Applications Performance Puzzles

Acknowledgments ix
 
Preface xi
 
1. Queuing Networks as Applications Models 1
 
1.1. Enterprise Applications--What Do They Have in Common?, 1
 
1.2. Key Performance Indicator--Transaction Time, 6
 
1.3. What Is Application Tuning and Sizing?, 8
 
1.4. Queuing Models of Enterprise Application, 9
 
1.5. Transaction Response Time and Transaction Profi le, 19
 
1.6. Network of Highways as an Analogy of the Queuing Model, 22
 
Take Away from the Chapter, 24
 
2. Building and Solving Application Models 25
 
2.1. Building Models, 25
 
Hardware Specification, 26
 
Model Topology, 28
 
A Model's Input Data, 29
 
Model Calibration, 31
 
2.2. Essentials of Queuing Networks Theory, 34
 
2.3. Solving Models, 39
 
2.4. Interpretation of Modeling Results, 47
 
Hardware Utilization, 47
 
Server Queue Length, Transaction Time, System Throughput, 51
 
Take Away from the Chapter, 54
 
3. Workload Characterization and Transaction Profiling 57
 
3.1. What Is Application Workload?, 57
 
3.2. Workload Characterization, 60
 
Transaction Rate and User Think Time, 61
 
Think Time Model, 65
 
Take Away from the Think Time Model, 68
 
Workload Deviations, 68
 
"Garbage in, Garbage out" Models, 68
 
Realistic Workload, 69
 
Users' Redistribution, 72
 
Changing Number of Users, 72
 
Transaction Rate Variation, 75
 
Take Away from "Garbage in, Garbage out" Models, 78
 
Number of Application Users, 78
 
User Concurrency Model, 80
 
Take Away from User Concurrency Model, 81
 
3.3. Business Process Analysis, 81
 
3.4. Mining Transactional Data from Production Applications, 88
 
Profi ling Transactions Using Operating System Monitors and Utilities, 88
 
Application Log Files, 90
 
Transaction Monitors,