20 practical guides

Cloud and DevOps Guides

Learn how to deploy and operate cloud systems with explicit failure boundaries, reviewed automation, useful telemetry, protected state, and tested recovery procedures.

Why this topic

Built for developers and operations teams deploying reliable systems who need concrete design checks and safe operating practices.

01
Cloud architecture fundamentals

Regions, zones, and availability

Design a cloud workload across failure boundaries, define its availability target, and test whether traffic and data survive a zone outage.

02
Cloud networking

Cloud network foundations

Plan address space, subnet boundaries, routes, and controlled connectivity so cloud workloads communicate without accidental exposure.

03
Application delivery

Traffic distribution layers

Route requests through health-aware front doors, preserve client context, and diagnose traffic behavior without exposing application instances.

04
Compute architecture

Containers and virtual machines

Compare isolation, packaging, startup, operations, and workload constraints before choosing containers, virtual machines, or both.

05
Container image engineering

Production Dockerfiles

Build reproducible container images with pinned inputs, multi-stage builds, non-root runtime users, clear startup behavior, and supply-chain evidence.

06
Developer environments

Local development with Compose

Define a safe multi-service local stack with health-aware startup, persistent test data, profiles, and clear reset procedures.

07
Kubernetes fundamentals

Core Kubernetes workloads

Connect Kubernetes workload units, rollout controllers, and stable network endpoints into one practical application deployment model.

08
Kubernetes configuration

Kubernetes application configuration

Separate ordinary configuration from confidential values, control access, trigger reliable rollouts, and verify how updates reach Pods.

09
Elastic capacity

Application autoscaling

Choose demand signals, capacity bounds, warm-up behavior, and load tests that let applications scale predictably without hiding bottlenecks.

10
Infrastructure automation

Terraform infrastructure as code

Model cloud resources declaratively, review plans, organize modules, and apply changes through a controlled Terraform workflow.

11
Terraform operations

Terraform state and backends

Protect Terraform state, prevent concurrent writes, separate environments, and recover state without unsafe manual editing.

12
Software delivery

CI/CD pipeline design

Build once, verify an immutable artifact, promote it through protected environments, and retain evidence plus a practiced recovery path.

13
Release strategies

Blue-green and canary releases

Choose between environment switching and gradual traffic exposure using measurable risk, capacity, state compatibility, and rollback needs.

14
Production observability

Logs, metrics, and traces

Instrument services with correlated events, time-series signals, and request traces that answer operational questions without leaking sensitive data.

15
Reliability management

Service level objectives

Turn user expectations into measurable service indicators, objectives, error budgets, and response rules that guide reliability work.

16
Cloud financial operations

Cloud cost optimization

Allocate cloud spend, find idle and oversized resources, evaluate commitments, and verify savings without weakening reliability or ownership.

17
Data resilience

Database backup and restore

Set recovery targets, create protected backups, restore into isolation, validate data and applications, and retain evidence that recovery works.

18
Operational resilience

Disaster recovery planning

Connect business priorities, recovery targets, dependency maps, recovery patterns, and exercises into an executable cloud disaster plan.

19
Production operations

Production configuration management

Classify settings, establish authoritative sources, validate changes, protect secrets, and roll production configuration through observable releases.

20
Performance troubleshooting

Diagnose cloud application latency

Bound a latency incident, follow one request across dependencies, test evidence-based hypotheses, and recover without destructive guesswork.

Demand and scope reference: AWS Well-Architected Reliability Pillar. Official AWS design principles and practices for reliable workloads and recovery planning.