Continuous Delivery Architecture study plan
Continuous Delivery Architecture · Give each week a clear focus.
8 weeks · About 7 hours per week
Deployment Pipeline Design and Orchestration & Continuous Delivery Foundations and Value Streams
Map a real or hypothetical value stream from commit to production, labeling every wait, handoff, and manual approval step and estimating lead time for each stage
Define deployment pipeline stages and quality gates for a sample service, distinguishing continuous integration, continuous delivery, and continuous deployment triggers
Compare blue-green, canary, and rolling pipeline orchestration strategies by listing the failure mode each one catches and the blast radius each one allows
Identify three bottlenecks in the mapped value stream and propose one automated control per bottleneck that removes a manual gate
Your checkpoint
A one-page value stream map plus a pipeline stage diagram (stages, gates, triggers, rollback point) for the sample service, with measured or estimated lead time per stage
Deployment Pipeline Design and Orchestration & Build and Dependency Management
Sequence a full pipeline (build, artifact publish, test tiers, environment promotion, deploy, verify) and justify the ordering of each stage
Select an artifact repository strategy including immutable versioning, retention, and promotion of the same binary across environments
Configure dependency pinning and lockfiles for one project and explain how reproducible builds prevent pipeline drift
Distinguish pipeline-as-code definitions from UI-configured jobs and describe how to version and review pipeline changes
Your checkpoint
A pipeline-as-code skeleton (or equivalent declarative definition) with at least five stages, artifact promotion rules, and dependency lockfile committed to a repository
Deployment Pipeline Design and Orchestration & Automated Testing Strategy in the Pipeline
Design a test pyramid for the sample service with explicit runtime budgets per tier and a rule for which tier owns which risk
Write the acceptance criteria for a smoke test suite and a post-deploy verification suite that run inside the pipeline
Define flaky-test triage policy including quarantine, ownership, and maximum allowed retry counts
Specify test data provisioning for pipeline runs so tests are deterministic and isolated
Your checkpoint
A test strategy document with tier table (scope, runtime budget, trigger, pass criteria), flaky-test policy, and test data approach
Automated Testing Strategy in the Pipeline & Environment Provisioning and Infrastructure as Code
Provision a repeatable environment from infrastructure-as-code definitions, including network, compute, and managed data store
Explain state management for infrastructure as code, including remote state, locking, and environment separation
Define environment parity requirements and identify where drift is acceptable versus prohibited
Add an ephemeral preview environment step to the pipeline design and specify its teardown trigger
Your checkpoint
Working infrastructure-as-code definitions that create and destroy one environment reproducibly, plus a drift and parity checklist
Environment Provisioning and Infrastructure as Code & Release and Deployment Patterns
Compare release versus deploy as separate concerns and describe how decoupling them enables low-risk delivery
Design a canary or blue-green release with automated metric-based promotion and automatic rollback thresholds
Specify feature-flag lifecycle rules including creation, rollout percentage, cleanup, and ownership
Define database-compatible rollout sequencing for a schema change that must survive both old and new application versions
Your checkpoint
A release plan for one feature covering deploy sequence, flag rollout schedule, rollback trigger thresholds, and backward-compatible migration steps
Configuration, Secrets, and Data Migration & Release and Deployment Patterns
Classify configuration by environment scope and separate build-time from runtime configuration
Design a secrets management approach covering storage, injection at runtime, rotation, and audit
Plan an expand-migrate-contract data migration with verification queries and a rollback path at each phase
Describe how to test configuration and secrets handling in the pipeline without exposing values in logs
Your checkpoint
A configuration and secrets matrix (setting, scope, source, rotation owner) plus a phased data migration runbook with verification and rollback steps
Observability, Feedback, and Governance & Continuous Delivery Foundations and Value Streams
Define the four delivery key metrics (lead time, deployment frequency, change failure rate, time to restore) and instrument each with a concrete data source
Design pipeline and production observability signals that feed automatic promotion and rollback decisions
Specify a post-incident review process that produces a pipeline or test change, not just a document
Draft governance controls (approval policy, audit trail, segregation of duties) that do not reintroduce manual gates into routine deploys
Your checkpoint
A metrics and observability dashboard spec with data sources, alert thresholds, and a governance control list mapped to pipeline stages
Continuous Delivery Foundations and Value Streams & Build and Dependency Management & Deployment Pipeline Design and Orchestration & Automated Testing Strategy in the Pipeline & Environment Provisioning and Infrastructure as Code & Release and Deployment Patterns & Configuration, Secrets, and Data Migration & Observability, Feedback, and Governance
Consolidate weak areas by scoring practice questions per domain and re-studying only domains below threshold
Re-derive the end-to-end architecture of a delivery system from memory: pipeline, environments, release pattern, config and secrets, observability
Explain trade-offs aloud for five scenarios (monolith to pipeline, regulated approval, multi-region release, legacy database, flaky suite) in under three minutes each
Build a personal one-page reference of decision rules and anti-patterns for final review
Your checkpoint
A timed full-length practice exam scored by domain, with a written remediation list and a one-page decision-rule reference sheet
Milestones
Week 2: Can draw the complete pipeline from memory with correct stage ordering, artifact promotion, and rollback point, and justify each stage in one sentence
Week 4: Can create and destroy an environment from code alone and explain state locking, drift detection, and the test tier runtime budgets without notes
Week 6: Can walk through a canary release with automatic rollback plus a backward-compatible schema migration end to end, naming the failure mode at each step
Week 8: Scores at least 80 percent overall on a timed practice exam with no domain below 70 percent, and can state the four key delivery metrics with their data sources