Service Quality of Cloud-Based Applications

ISBN-13: 9781118763292 / Angielski / Twarda / 2014 / 344 str.

Eric Bauer; Randee Adams

Opis

This book explains why applications running on cloud might not deliver the same service reliability, availability, latency and overall quality to end users as they do when the applications are running on traditional (non-virtualized, non-cloud) configurations, and explains what can be done to mitigate that risk.

Spis treści

Figures xv

Tables and Equations xxi

1 INTRODUCTION 1

1.1 Approach 1

1.2 Target Audience 3

1.3 Organization 3

I CONTEXT 7

2 APPLICATION SERVICE QUALITY 9

2.1 Simple Application Model 9

2.2 Service Boundaries 11

2.3 Key Quality and Performance Indicators 12

2.4 Key Application Characteristics 15

2.5 Application Service Quality Metrics 17

2.6 Technical Service versus Support Service 27

2.7 Security Considerations 28

3 CLOUD MODEL 29

3.1 Roles in Cloud Computing 30

3.2 Cloud Service Models 30

3.3 Cloud Essential Characteristics 31

3.4 Simplifi ed Cloud Architecture 33

3.5 Elasticity Measurements 36

3.6 Regions and Zones 44

3.7 Cloud Awareness 45

4 VIRTUALIZED INFRASTRUCTURE IMPAIRMENTS 49

4.1 Service Latency, Virtualization, and the Cloud 50

4.2 VM Failure 54

4.3 Nondelivery of Configured VM Capacity 54

4.4 Delivery of Degraded VM Capacity 57

4.5 Tail Latency 59

4.6 Clock Event Jitter 60

4.7 Clock Drift 61

4.8 Failed or Slow Allocation and Startup of VM Instance 62

4.9 Outlook for Virtualized Infrastructure Impairments 63

II ANALYSIS 65

5 APPLICATION REDUNDANCY AND CLOUD COMPUTING 67

5.1 Failures, Availability, and Simplex Architectures 68

5.2 Improving Software Repair Times via Virtualization 70

5.3 Improving Infrastructure Repair Times via Virtualization 72

5.4 Redundancy and Recoverability 75

5.5 Sequential Redundancy and Concurrent Redundancy 80

5.6 Application Service Impact of Virtualization Impairments 84

5.7 Data Redundancy 90

5.8 Discussion 92

6 LOAD DISTRIBUTION AND BALANCING 97

6.1 Load Distribution Mechanisms 97

6.2 Load Distribution Strategies 99

6.3 Proxy Load Balancers 99

6.4 Nonproxy Load Distribution 101

6.5 Hierarchy of Load Distribution 102

6.6 Cloud–Based Load Balancing Challenges 103

6.7 The Role of Load Balancing in Support of Redundancy 103

6.8 Load Balancing and Availability Zones 104

6.9 Workload Service Measurements 104

6.10 Operational Considerations 105

6.11 Load Balancing and Application Service Quality 107

7 FAILURE CONTAINMENT 111

7.1 Failure Containment 111

7.2 Points of Failure 116

7.3 Extreme Solution Coresidency 122

7.4 Multitenancy and Solution Containers 124

8 CAPACITY MANAGEMENT 127

8.1 Workload Variations 128

8.2 Traditional Capacity Management 129

8.3 Traditional Overload Control 129

8.4 Capacity Management and Virtualization 131

8.5 Capacity Management in Cloud 133

8.6 Storage Elasticity Considerations 135

8.7 Elasticity and Overload 136

8.8 Operational Considerations 137

8.9 Workload Whipsaw 138

8.10 General Elasticity Risks 140

8.11 Elasticity Failure Scenarios 141

9 RELEASE MANAGEMENT 145

9.1 Terminology 145

9.2 Traditional Software Upgrade Strategies 146

9.3 Cloud–Enabled Software Upgrade Strategies 153

9.4 Data Management 158

9.5 Role of Service Orchestration in Software Upgrade 159

9.6 Conclusion 161

10 END–TO–END CONSIDERATIONS 163

10.1 End–to–End Service Context 163

10.2 Three–Layer End–to–End Service Model 169

10.3 Distributed and Centralized Cloud Data Centers 177

10.4 Multitiered Solution Architectures 183

10.5 Disaster Recovery and Geographic Redundancy 184

III RECOMMENDATIONS 191

11 ACCOUNTABILITIES FOR SERVICE QUALITY 193

11.1 Traditional Accountability 193

11.2 The Cloud Service Delivery Path 194

11.3 Cloud Accountability 197

11.4 Accountability Case Studies 200

11.5 Service Quality Gap Model 205

11.6 Service Level Agreements 210

12 SERVICE AVAILABILITY MEASUREMENT 213

12.1 Parsimonious Service Measurements 214

12.2 Traditional Service Availability Measurement 215

12.3 Evolving Service Availability Measurements 217

12.4 Evolving Hardware Reliability Measurement 226

12.5 Evolving Elasticity Service Availability Measurements 228

12.6 Evolving Release Management Service Availability Measurement 229

12.7 Service Measurement Outlook 231

13 APPLICATION SERVICE QUALITY REQUIREMENTS 233

13.1 Service Availability Requirements 234

13.2 Service Latency Requirements 237

13.3 Service Reliability Requirements 237

13.4 Service Accessibility Requirements 238

13.5 Service Retainability Requirements 239

13.6 Service Throughput Requirements 239

13.7 Timestamp Accuracy Requirements 240

13.8 Elasticity Requirements 240

13.9 Release Management Requirements 241

13.10 Disaster Recovery Requirements 241

14 VIRTUALIZED INFRASTRUCTURE MEASUREMENT AND MANAGEMENT 243

14.1 Business Context for Infrastructure Service Quality Measurements 244

14.2 Cloud Consumer Measurement Options 245

14.3 Impairment Measurement Strategies 247

14.4 Managing Virtualized Infrastructure Impairments 252

15 ANALYSIS OF CLOUD–BASED APPLICATIONS 255

15.1 Reliability Block Diagrams and Side–by–Side Analysis 256

15.2 IaaS Impairment Effects Analysis 257

15.3 PaaS Failure Effects Analysis 259

15.4 Workload Distribution Analysis 260

15.5 Anti–Affi nity Analysis 262

15.6 Elasticity Analysis 263

15.7 Release Management Impact Effects Analysis 267

15.8 Recovery Point Objective Analysis 268

15.9 Recovery Time Objective Analysis 270

16 TESTING CONSIDERATIONS 273

16.1 Context for Testing 273

16.2 Test Strategy 274

16.3 Simulating Infrastructure Impairments 277

16.4 Test Planning 278

17 CONNECTING THE DOTS 287

17.1 The Application Service Quality Challenge 287

17.2 Redundancy and Robustness 289

17.3 Design for Scalability 292

17.4 Design for Extensibility 292

17.5 Design for Failure 293

17.6 Planning Considerations 294

17.7 Evolving Traditional Applications 296

17.8 Concluding Remarks 301

Abbreviations 303

References 307

About the Authors 311

Index 313

O autorze

ERIC BAUER, MS, is Reliability Engineering Manager in the IP Platforms CTO of Alcatel–Lucent. The holder of more than a dozen U.S. patents, Mr. Bauer is the author or coauthor of Reliability and Availability of Cloud Computing, Beyond Redundancy: How Geographic Redundancy Can Improve Service Availability and Reliability of Computer–Based Systems, Design for Reliability: Information and Computer–Based Systems, and Practical System Reliability.

RANDEE ADAMS, MS, is a Consulting Member of the Technical Staff in the IP Platforms CTO of Alcatel–Lucent. She is the coauthor of Beyond Redundancy: How Geographic Redundancy Can Improve Service Availability and Reliability of Computer–Based Systems and Reliability and Availability of Cloud Computing. Ms. Adams has worked on many projects, including software development, delivery, engineering, architecture, and design.

Blurb

Learn how to make the most of cloud computing infrastructures

We expect that software applications and services delivered via cloud computing infrastructures will deliver the same level of service quality, reliability, and availability as they do when running on traditional native hardware configurations. Cloud computing infrastructures, however, introduce many potential service impairment risks based on their virtualized computing, memory, storage, and networking resources. As a result, cloud–based application developers must learn to mitigate these impairments to ensure acceptable service delivered to end users.

Service Quality of Cloud–Based Applications methodically analyzes the effect of cloud infrastructure impairments on services delivered to end users as well as the opportunities to improve service afforded by the cloud. The book recommends cloud architectures, policies, and techniques to maximize the ability of providing superior service.

Expertly written and organized, this book is divided into three parts:

• Part I: Context sets the foundation with discussion of application service quality, the cloud model, and virtualized infrastructure impairments.
• Part II: Analysis methodically considers how application service is affected by infrastructure impairments.
• Part III: Recommendations sets forth strategies and techniques to mitigate or eliminate the service quality risks and impairments of both current cloud–based applications and those in development.

An extensive list of references at the end of the book enables readers to explore individual topics in greater depth.

With its expert advice on architecting and engineering cloud–based applications, Service Quality of Cloud–Based Applications makes it possible for application architects, developers, and testers to meet their customers and end users needs and expectations.