DevOps and Cloud Concepts

DevOps is a combination of cultural philosophies, practices, and tools that increases an organization’s ability to deliver applications and services at high velocity. The term comes from merging “development” (Dev) and “operations” (Ops). Traditionally, software development and IT operations were separate teams with conflicting goals: developers wanted to release new features quickly, while operations wanted stability and uptime. This caused friction, slow release cycles, and frequent deployment failures. DevOps breaks down these silos by promoting collaboration, automation, and monitoring across the entire application lifecycle—from planning and coding to testing, deployment, and operations.

For AZ-900, you need to know that DevOps is not a specific Azure service but a methodology that Azure supports through various tools. The exam may ask why an organization would adopt DevOps (e.g., faster time to market, improved reliability, better collaboration) and which Azure services enable DevOps practices (e.g., Azure DevOps, GitHub Actions, Azure Boards, Azure Pipelines).

Azure DevOps is a suite of services that provides end-to-end DevOps tooling. The exam expects you to recognize the main components:

Additionally, GitHub Actions is another CI/CD tool that is tightly integrated with Azure. GitHub Actions allows you to automate workflows directly from your GitHub repository, and it can deploy to Azure services like App Service, Functions, and VMs.

CI is the practice of merging code changes into a shared mainline multiple times a day. Each integration is verified by an automated build and test process. Here’s the step-by-step mechanism:

Key benefits: Reduces integration problems, improves code quality, and provides rapid feedback.

CD extends CI by automatically deploying all code changes to a testing or staging environment after the build stage. Continuous deployment goes further by automatically deploying to production. The typical CD pipeline:

Azure Pipelines supports multi-stage YAML pipelines that define the entire CI/CD flow in code, enabling infrastructure as code for pipelines themselves.

DevOps often goes hand-in-hand with IaC, where you manage and provision cloud resources through machine-readable definition files rather than manual configuration. Azure supports:

With IaC, you can version control your infrastructure, automate deployments, and ensure consistency across environments. For example, a single Bicep file can define a virtual network, subnets, VMs, and network security groups, and be deployed via Azure Pipelines.

In an on-premises environment, teams typically used tools like Jenkins, TeamCity, or self-hosted Git servers. The operations team manually configured servers, installed dependencies, and deployed code via scripts or copy-paste. This was slow, error-prone, and required significant manual effort. Azure DevOps provides a managed, cloud-hosted alternative that scales automatically, integrates with other Azure services, and offers built-in security and compliance. You don’t need to manage build agents or infrastructure; Azure Pipelines provides hosted agents with pre-installed tools.

To get started with Azure DevOps:

Example CLI commands:

# Login to Azure DevOps
az devops login --organization https://dev.azure.com/contoso

# Create a new project
az devops project create --name MyProject --process Agile

# List pipelines
az pipelines list --project MyProject

Alternatively, you can use the Azure CLI with the devops extension:

az extension add --name azure-devops
az devops configure --defaults organization=https://dev.azure.com/contoso project=MyProject

Cloud computing accelerates DevOps by providing on-demand resources, self-service capabilities, and pay-as-you-go pricing. Azure’s global infrastructure allows you to spin up test environments quickly, scale production apps automatically, and monitor everything with built-in tools. The shared responsibility model also applies: while Azure secures the physical infrastructure, you are responsible for securing your code, pipelines, and secrets (e.g., using Azure Key Vault).

For AZ-900, remember that DevOps is a key reason organizations move to the cloud—it enables faster innovation and more reliable releases.

A mid-sized e-commerce company wants to release new features every week instead of every quarter. Their legacy process involved manual builds, copy-paste deployments, and a dedicated ops team that worked overnight. They adopt Azure DevOps with GitHub for source control. The development team creates feature branches, opens pull requests, and automated builds run unit tests and code analysis. After approval, the code merges to main, triggering a CI pipeline that builds and runs integration tests. The CD pipeline then deploys to a staging environment where QA performs exploratory testing using Azure Test Plans. Once approved, a manual release to production uses a blue-green deployment strategy with Azure App Service deployment slots. This reduces release time from weeks to hours and catches bugs early. However, they initially misconfigured the YAML pipeline to deploy to production on every push, causing an outage. They learned to add environment approvals and deployment gates.

A bank needs to release regulatory updates quickly but must adhere to strict audit trails and separation of duties. They use Azure DevOps with Azure Boards to track work items linked to compliance requirements. All code changes require a pull request with at least two reviewers and passing security scans (SonarQube, OWASP). The CI pipeline runs static code analysis and dependency scanning. The CD pipeline deploys to a pre-production environment that mirrors production, where automated regression tests run. Deployment to production requires a manual approval from a security officer. All pipeline runs are logged, and Azure Artifacts store immutable build outputs with retention policies. The bank uses Azure Policy to enforce resource tagging and compliance. They also use Azure Key Vault to manage secrets, avoiding hard-coded credentials in the pipeline. A common pitfall is not rotating service connection secrets regularly, which they address with automated secret rotation.

A small startup with three developers wants to deploy a containerized microservices app to Azure Kubernetes Service (AKS). They have no dedicated operations team. They use GitHub Actions for CI/CD because it integrates seamlessly with their GitHub repos. Their workflow builds Docker images, pushes them to Azure Container Registry, and updates the AKS deployment using kubectl. They use Azure Monitor and Application Insights to get alerts on app performance. The startup saves costs by using Azure Pipelines free tier (1800 minutes/month) and AKS dev/test pricing. A common mistake is not setting up proper resource limits on containers, causing a single service to consume all cluster resources. They learned to define resource requests and limits in the Kubernetes manifests. The DevOps culture enables them to focus on features rather than infrastructure.