book

Production-Ready Microservices

by Susan J. Fowler

November 2016

Intermediate to advanced

172 pages

5h 5m

English

O'Reilly Media, Inc.

Read now

Unlock full access

Who This Book Is Written ForWhat This Book Is NotHow To Use This BookHow This Book Is StructuredConventions Used in This BookO’Reilly SafariHow to Contact UsAcknowledgments
From Monoliths to MicroservicesMicroservice ArchitectureThe Microservice EcosystemLayer 1: HardwareLayer 2: CommunicationLayer 3: The Application PlatformLayer 4: MicroservicesOrganizational ChallengesThe Inverse Conway’s LawTechnical SprawlMore Ways to FailCompetition for Resources
The Challenges of Microservice StandardizationAvailability: The Goal of StandardizationProduction-Readiness StandardsStabilityReliabilityScalabilityFault Tolerance and Catastrophe-PreparednessPerformanceMonitoringDocumentationImplementing Production-Readiness
Principles of Building Stable and Reliable MicroservicesThe Development CycleThe Deployment PipelineStagingCanaryProductionEnforcing Stable and Reliable DeploymentDependenciesRouting and DiscoveryDeprecation and DecommissioningEvaluate Your MicroserviceThe Development CycleThe Deployment PipelineDependenciesRouting and DiscoveryDeprecation and Decommissioning
Principles of Microservice Scalability and PerformanceKnowing the Growth ScaleThe Qualitative Growth ScaleThe Quantitative Growth ScaleEfficient Use of ResourcesResource AwarenessResource RequirementsResource BottlenecksCapacity PlanningDependency ScalingTraffic ManagementTask Handling and ProcessingProgramming Language LimitationsHandling Requests and Processing Tasks EfficientlyScalable Data StorageDatabase Choice in Microservice EcosystemsDatabase Challenges in Microservice ArchitectureEvaluate Your MicroserviceKnowing the Growth ScaleEfficient Use of ResourcesResource AwarenessCapacity PlanningDependency ScalingTraffic ManagementTask Handling and ProcessingScalable Data Storage
Principles of Building Fault-Tolerant MicroservicesAvoiding Single Points of FailureCatastrophes and Failure ScenariosCommon Failures Across an EcosystemHardware FailuresCommunication-Level and Application Platform–Level FailuresDependency FailuresInternal (Microservice) FailuresResiliency TestingCode TestingLoad TestingChaos TestingFailure Detection and RemediationIncidents and OutagesAppropriate CategorizationThe Five Stages of Incident ResponseEvaluate Your MicroserviceAvoiding Single Points of FailureCatastrophes and Failure ScenariosResiliency TestingFailure Detection and Remediation
Principles of Microservice MonitoringKey MetricsLoggingDashboardsAlertingSetting up Effective AlertingHandling AlertsOn-Call RotationsEvaluate Your MicroserviceKey MetricsLoggingDashboardsAlertingOn-Call Rotations
Principles of Microservice Documentation and UnderstandingMicroservice DocumentationDescriptionArchitecture DiagramContact and On-Call InformationLinksOnboarding and Development GuideRequest Flows, Endpoints, and DependenciesOn-Call RunbooksFAQMicroservice UnderstandingArchitecture ReviewsProduction-Readiness AuditsProduction-Readiness RoadmapsProduction-Readiness AutomationEvaluate Your MicroserviceMicroservice DocumentationMicroservice Understanding
A Production-Ready Service Is Stable and ReliableA Production-Ready Service Is Scalable and PerformantA Production-Ready Service Is Fault Tolerant and Prepared for Any CatastropheA Production-Ready Service Is Properly MonitoredA Production-Ready Service Is Documented and Understood
Stability and ReliabilityThe Development CycleThe Deployment PipelineDependenciesRouting and DiscoveryDeprecation and DecommissioningScalability and PerformanceKnowing the Growth ScaleEfficient Use of ResourcesResource AwarenessCapacity PlanningDependency ScalingTraffic ManagementTask Handling and ProcessingScalable Data StorageFault Tolerance and Catastrophe-PreparednessAvoiding Single Points of FailureCatastrophes and Failure ScenariosResiliency TestingFailure Detection and RemediationMonitoringKey MetricsLoggingDashboardsAlertingOn-Call RotationsDocumentation and UnderstandingMicroservice DocumentationMicroservice Understanding

Content preview from Production-Ready Microservices

Chapter 3. Stability and Reliability

A production-ready microservice is stable and reliable. Both individual microservices and the overall microservice ecosystem are constantly changing and evolving, and any efforts made to increase the stability and reliability of a microservice go a long way toward ensuring the health and availability of the overall ecosystem. In this chapter, different ways to build and run a stable and reliable microservice are explored, including standardizing the development process, building comprehensive deployment pipelines, understanding dependencies and protecting against their failures, building stable and reliable routing and discovery, and establishing appropriate deprecation and decommissioning procedures for old or outdated microservices and/or their endpoints.

Principles of Building Stable and Reliable Microservices

Microservice architecture lends itself to fast-paced development. The freedom offered by microservices means that the ecosystem will be in a state of continuous change, never static, always evolving. Features will be added every day, new builds will be deployed multiple times per day, and old technologies will be swapped for newer and better ones at an astounding pace. This freedom and flexibility gives rise to real, tangible innovation, but comes at a great cost.

Innovation, increased developer velocity and productivity, rapid technological advancement, and the ever-changing microservice ecosystem can all very quickly be brought ...