book

Cloud Native Infrastructure

by Justin Garrison, Kris Nova

November 2017

Intermediate to advanced

160 pages

4h 13m

English

O'Reilly Media, Inc.

Read now

Unlock full access

Who Should Read This BookWhy We Wrote This BookNavigating This BookOnline ResourcesConventions Used in This BookO’Reilly SafariHow to Contact UsAcknowledgmentsJustin GarrisonKris Nova
Cloud Native BenefitsServersVirtualizationInfrastructure as a ServicePlatform as a ServiceCloud Native InfrastructureWhat Is Not Cloud Native Infrastructure?Cloud Native ApplicationsMicroservicesHealth ReportingTelemetry DataResiliencyDeclarative, Not ReactiveHow Do Cloud Native Applications Impact Infrastructure?Conclusion
ApplicationsPeopleSystemsBusinessWhen You Don’t Need Cloud Native InfrastructureTechnical LimitationsBusiness LimitationsConclusion
Representing InfrastructureInfrastructure as a DiagramInfrastructure as a ScriptInfrastructure as CodeInfrastructure as SoftwareDeployment ToolsIdempotencyHandling FailureConclusion
The Bootstrapping ProblemThe APIThe State of the WorldThe Reconciler PatternRule 1: Use a Data Structure for All Inputs and OutputsRule 2: Ensure That the Data Structure Is ImmutableRule 3: Keep the Resource Map SimpleRule 4: Make the Actual State Match the Expected StateThe Reconciler Pattern’s MethodsExample of the Pattern in GoThe Auditing RelationshipUsing the Reconciler Pattern in a ControllerConclusion
Designing an APIAdding FeaturesDeprecating FeaturesMutating InfrastructureConclusion
What Are We Testing?Writing Testable CodeValidationEntering Your CodebaseSelf-AwarenessTypes of TestsInfrastructure AssertionsIntegration TestingUnit TestingMock TestingChaos TestingMonitoring InfrastructureConclusion
Application DesignImplementing Cloud Native PatternsApplication Life CycleDeployRunRetireApplication Requirements on InfrastructureApplication Runtime and IsolationResource Allocation and SchedulingEnvironment IsolationService DiscoveryState ManagementMonitoring and LoggingMetrics AggregationDebugging and TracingConclusion
Policy as CodeDeployment GatingConformity TestingCompliance TestingActivity TestingAuditing InfrastructureImmutable InfrastructureConclusion
Where to Focus for ChangePeopleArchitectureChaos ManagementApplicationsPredicting the FutureConclusion

Load BalancingLoad SheddingService DiscoveryRetries and DeadlinesCircuit BreakingTLS and AuthRouting (Ingress and Egress)Insight and Monitoring
Lock-in Is UnavoidableTechnology Lock-inVendor Lock-inLock-in Is a RiskDon’t Outsource Thinking

Content preview from Cloud Native Infrastructure

Chapter 3. Evolution of Cloud Native Deployments

We discussed in the previous chapter what the requirements are before you adopt cloud native infrastructure. Having API-driven infrastructure provisioning (IaaS) will be a requirement before you deploy.

In this chapter, we explore taking the idea of cloud native infrastructure topologies and realizing them in a cloud. We’ll learn common tools and patterns that help operators control their infrastructure.

The first step in deploying infrastructure is being able to represent it. Traditionally this could have been handled on a whiteboard, or, if you’re lucky, in documentation stored on a company wiki. Today, things are moving into a more programmatic space, and infrastructure representation is commonly documented in ways that make it easy for applications to interpret it. Regardless of the avenue in which it is represented, the need to comprehensively represent the infrastructure remains the same.

As one would expect, defining infrastructure in the cloud can range from very simple designs to very complex designs. Regardless of the complexity, great attention to detail must be given to infrastructure representation to ensure the design is reproducible. Being able to clearly transmit ideas is even more important. Therefore, having a clear, accurate, and easy-to-understand representation of infrastructure-level resources is imperative.

We also gain a lot from having a well-crafted representation: