Operating Continuously: Best Practices for Accelerating Software Delivery

https://mhco.ca/scegr9cg6 Preface Who Will Benefit from Reading This Book? Navigating and Using This Book Conventions Used in This Book O’Reilly Online Learning How to Contact Us Acknowledgments Introduction It’s Never Been Just About Dev and Ops The Steps Need a Refresh The Steps Happen in Parallel 1. Deploy Redefining Release and Deploy What Deploy and Release Used to Mean What Deploy and Release Mean Now Small and Frequent > Large and Infrequent Deploy Can Be Decoupled from Release Merge Conflicts and Their Impact on Deployments Characteristics of a High-Performance Deployment System Visibility Deployment Orchestration Deployment Strategies and Shipping Code Blue/Green Deployments and Canaries Deploying Code DORA Metrics (Minus MTTR) Summary 2. Release Feature Flags The Power of Feature Flags Granular Targeting Progressive Delivery Testing in Production Personalization Entitlements Feature Management Platforms Speed Experimentation Security Reliability Compliance Delegation Progression Flag Management Languages Workflow Enhancements Summary 3. Operate What Is an Incident? Common Incident Causes Changes to Your System Changes to the Inputs of Your System Nonsoftware Issues Identifying and Declaring an Incident Automatically Manually Mitigating Risk Rollbacks Kill Switches Safety Valves Handling Migrations Observability Principles of Healthy Incident Management Self-Identification Continuous Improvement Runbook Operability High Volume, Low Impact What Bad Incident Management Looks Like What Good Incident Management Looks Like What Makes an Incident Response Process Successful? People Roles Mandatory Versus Optional Culture Tools Process Post-Incident Review Metrics Response Time Summary 4. Measure and Experiment Measure Why We Need to Measure Data What Impact Does Measure Have? Shifting from Measure to Experiment Experimentation New Software Delivery Enables Experimentation Feature Validation Experiments Incentive-Based Development Pitfalls of Changing Success Metrics A New Product in a New Market Facing the Results Culture and Feature Abandonment Cheap Features A/B/n Testing Designing a Test Running the Test Choosing the Right Experimentation Platform Risk Mitigation Defining Risk Mitigation Measuring Risk Experimentation for Risk Mitigation Optimization Experimentation Examples Measuring a Sign-Up Flow Fifty Shades of Blue Summary Conclusion Rethinking the Loop Deploy Release Operate Measure and Experiment Final Words Index About the Authors