Performance Evaluation for Network Services, Systems and Protocols

This book provides a comprehensive view of the methods and approaches for performance evaluation of computer networks. It offers a clear and logical introduction to the topic, covering both fundamental concepts and practical aspects. It enables the reader to answer a series of questions regarding performance evaluation in modern computer networking scenarios, such as ‘What, where, and when to measure?’, ‘Which time scale is more appropriate for a particular measurement and analysis?’, 'Experimentation, simulation or emulation? Why?’, and ‘How do I best design a sound performance evaluation plan?’. The book includes concrete examples and applications in the important aspects of experimentation, simulation and emulation, and analytical modeling, with strong support from the scientific literature. It enables the identification of common shortcomings and highlights where students, researchers, and engineers should focus to conduct sound performance evaluation. This book is a useful guide to advanced undergraduates and graduate students, network engineers, and researchers who plan and design proper performance evaluation of computer networks and services. Previous knowledge of computer networks concepts, mechanisms, and protocols is assumed. Although the book provides a quick review on applied statistics in computer networking, familiarity with basic statistics is an asset. It is suitable for advanced courses on computer networking as well as for more specific courses as a secondary textbook. Foreword 5 Acknowledgments 7 Contents 9 Chapter 1: Principles of Performance Evaluation of Computer Networks 12 1.1 Motivation: Why Do We Need to Assess the Performance of Computer Networks? 12 1.2 Classical and Modern Scenarios: Examples from Research Papers 13 1.2.1 Performance Evaluation in Classical Scenarios 14 1.2.1.1 Application Layer 14 1.2.1.2 Transport Layer 18 1.2.1.3 Network Layer 22 1.2.1.4 Link Layer 25 1.2.2 Performance Evaluation in Modern Scenarios 27 1.2.2.1 Virtualization and Cloud Computing 27 1.2.2.2 Software-Defined Networking 33 1.2.2.3 Network Functions Virtualization 38 1.3 The Pillars of Performance Evaluation of Networking and Communication Systems 41 1.3.1 Experimentation/Prototyping, Simulation/Emulation, and Modeling 41 1.3.1.1 Network Experimentation and Prototyping 42 1.3.1.2 Network Simulation and Emulation 43 1.3.1.3 Analytical Modeling 45 Video Game Characterization 46 TCP Throughput 46 Aggregate Background Traffic 48 Error Models for Communication Channels 49 1.3.2 Supporting Strategies: Measurements 50 References 51 Chapter 2: Methods and Techniques for Measurements in the Internet 55 2.1 Passive vs. Active vs. Hybrid Measurements 57 2.2 Traffic Measurements: Packets, Flow Records, and Aggregated Data 61 2.3 Sampling Techniques for Network Management 64 2.4 Internet Topology: Measurements, Modeling, and Analysis 67 2.4.1 Internet Topology Resolution 69 2.4.2 Internet Topology Discovery: Tools, Techniques, and Datasets 71 2.4.2.1 Topology Discovery Tools 71 2.4.2.2 Topology Generator Tools 72 2.4.2.3 Topology Datasets 72 2.5 Challenges for Traffic Measurements and Analyses in Virtual Environments 74 2.5.1 Cloud Computing Environments 74 2.5.2 Virtualization at Network Level 77 2.6 Bandwidth Estimation Methods 79 References 80 Chapter 3: A Primer on Applied Statistics in Computer Networking 84 3.1 Statistics and Computational Statistics 84 3.2 I’m All About That Data 87 3.3 Essential Concepts and Terminology 89 3.4 Descriptive Statistics 91 3.4.1 I Mean It (Or Measures of Centrality) 92 3.4.2 This Is Dull (Or Measures of Dispersion) 92 3.4.3 Is It Paranormally Distributed? (Or Measures of Asymmetry and Tailedness) (Fig. 3.2) 95 3.5 Inferential Statistics 98 3.5.1 Parameter Estimation: Point vs. Interval 99 3.5.1.1 Confidence Intervals 100 Confidence Interval for the Mean 101 3.5.2 Estimators and Estimation Methods 104 3.5.2.1 Essential Properties for Estimators 105 3.5.2.2 Likelihood Function and Its Estimation 105 3.6 The Heavy-Tailed Phenomenon 108 3.6.1 Outlier Detection 109 3.6.2 Heavy-Tailed Distributions and Its Variations (Subclasses) 111 3.6.2.1 What Is This That Stands Before Me? 112 3.6.2.2 A Less Informal Look at the Tail 114 Right Tail Properties of Distributions 114 3.6.3 Evidence of Heavy-Tailedness in Computer Networks 116 References 120 Chapter 4: Internet Traffic Profiling 122 4.1 Traffic Analysis 123 4.1.1 Identification and Classification 123 4.1.2 Techniques, Tools, and Systems for Traffic Profiling 127 4.1.2.1 Port-Based Traffic Profiling 127 4.1.2.2 Shallow and Deep Packet Inspection 128 DPI in Practice 131 Specialized Architectures and Hardware Accelerators 131 Specialized Libraries and Operating System Support 134 4.1.2.3 Flow-Based Traffic Profiling 136 Applied Machine Learning for Internet Traffic Profiling 138 4.2 Industrial Approach for Traffic Profiling: Products and Services 139 4.3 Traffic Models in Practice 141 4.3.1 Workload Generators 142 4.4 Simulation and Emulation 146 4.4.1 Discrete-Event Simulation and Network Simulation Environments 149 4.4.2 Practical Use of Network Simulators and Traffic Profiles 156 References 157 Chapter 5: Designing and Executing Experimental Plans 162 5.1 Designing Performance Evaluation Plans: Fundamentals 162 5.2 Design of Experiments (DoE) 165 5.2.1 The DoE Jargon 168 5.2.2 To Replicate or to Slice? 170 5.3 DOE Options: Choosing a Proper Design 173 5.3.1 Classification of DOE Methods 174 5.3.2 Notation 175 5.4 Experimental Designs 175 5.4.1 2k Factorial Designs (a.k.a. Coarse Grids) 175 5.4.2 2k − p Fractional Factorial Designs 176 5.4.3 mk Factorial Designs (a.k.a. Finer Grids) 176 5.5 Test, Validation, Analysis, and Interpretation of DOE Results 176 5.6 DOEs: Some Pitfalls and Caveats 177 5.7 DOE in Computer Networking Problems 178 5.7.1 General Guidelines 178 5.7.1.1 Advice for Conducting Experiments in Computer Networking 182 5.7.2 Hands-On 183 References 183 Front Matter....Pages i-xi Principles of Performance Evaluation of Computer Networks....Pages 1-43 Methods and Techniques for Measurements in the Internet....Pages 45-73 A Primer on Applied Statistics in Computer Networking....Pages 75-112 Internet Traffic Profiling....Pages 113-152 Designing and Executing Experimental Plans....Pages 153-175