| Deployment frequency |
On-demand (multiple times per day) |
Between once per day and once per week |
Between once per week and once per month |
Between once per month and once every six months |
| Lead time for changes |
Less than one hour |
Between one day and one week |
Between one week and one month |
Between one month and six months |
| Change failure rate |
0-15% |
16-30% |
16-30% |
46-60% |
| Time to restore service |
Less than one hour |
Less than one day |
Between one day and one week |
More than six months |
Source: DORA State of DevOps Report, 2024.
These metrics are not vanity numbers. The DORA research demonstrates that elite performers deliver better business outcomes: higher profitability, greater market share, and higher employee satisfaction. Measuring these metrics is the first step toward improving them.
Core DevOps practices
Continuous Integration (CI)
Developers merge code changes into a shared repository frequently -- at minimum once per day. Each merge triggers an automated build and test suite. If tests fail, the team fixes the issue immediately rather than letting it accumulate.
Why it matters: Without CI, teams integrate code every few weeks or months, leading to painful "merge hell" where conflicting changes take days to resolve. CI catches integration issues within minutes.
Tools: GitHub Actions, GitLab CI/CD, Jenkins, CircleCI, Azure DevOps Pipelines, AWS CodeBuild.
Continuous Delivery and Continuous Deployment (CD)
Continuous delivery means every code change that passes automated tests is automatically prepared for release to production. A human approves the final deployment.
Continuous deployment goes further: every change that passes tests deploys to production automatically, with no human approval step.
Why it matters: CD reduces the risk of each deployment by making deployments small and frequent. Deploying 10 changes at once is riskier than deploying 10 changes one at a time because it is harder to identify which change caused a problem.
Tools: ArgoCD, Flux, Spinnaker, AWS CodeDeploy, Azure DevOps Release Pipelines, GitHub Actions.
Infrastructure as Code (IaC)
Managing infrastructure through version-controlled configuration files rather than manual processes. See the companion article: What Is Infrastructure as Code?
Tools: Terraform, AWS CloudFormation, Pulumi, Ansible, Azure Bicep.
Monitoring and Observability
Collecting metrics, logs, and traces from applications and infrastructure to understand system behavior, detect anomalies, and diagnose problems.
Monitoring answers "is the system working?" Observability answers "why is the system not working?" Observability requires instrumented code that produces telemetry (metrics, structured logs, distributed traces) allowing engineers to ask arbitrary questions about system behavior.
Tools: Prometheus, Grafana, Datadog, New Relic, AWS CloudWatch, Elastic Stack (ELK), Jaeger, OpenTelemetry.
Configuration Management
Ensuring all servers and environments are configured consistently and automatically. Configuration management tools enforce the desired state of software, packages, files, and services across fleets of servers.
Tools: Ansible, Chef, Puppet, SaltStack.
Containerization and Orchestration
Packaging applications with their dependencies into containers that run consistently across environments. Container orchestration platforms manage the lifecycle of containers at scale.
Tools: Docker, Kubernetes, Amazon ECS, Azure Container Instances, Google Cloud Run.
DevOps, DevSecOps, and Platform Engineering
DevSecOps
DevSecOps integrates security practices into every stage of the DevOps pipeline rather than treating security as a separate phase at the end. Security scanning happens automatically in CI/CD: static analysis (SAST), dynamic analysis (DAST), dependency vulnerability scanning (SCA), container image scanning, and Infrastructure as Code policy checks.
The motivation is simple: fixing a security vulnerability in development costs 10x less than fixing it in production. Shifting security left (earlier in the pipeline) reduces cost and risk.
Platform Engineering
Platform engineering emerged from the observation that DevOps, in practice, often overloaded development teams with operational responsibilities. Platform engineering creates internal developer platforms (IDPs) that abstract away infrastructure complexity, giving developers self-service tools for provisioning environments, deploying applications, and monitoring services.
In 2026, platform engineering is the fastest-growing discipline adjacent to DevOps. Gartner predicts that 80% of large software engineering organizations will have platform engineering teams by 2027.
GitOps
GitOps applies DevOps best practices (version control, code review, CI/CD) specifically to infrastructure and operational workflows. The Git repository becomes the single source of truth for both application code and infrastructure state. Tools like ArgoCD and Flux continuously reconcile the actual state of a Kubernetes cluster with the desired state declared in Git.
The DevOps toolchain
A complete DevOps toolchain covers the entire software delivery lifecycle:
| Stage |
Purpose |
Common tools |
| Plan |
Requirements, user stories, sprint planning |
Jira, Linear, GitHub Issues, Azure Boards |
| Code |
Writing and reviewing code |
VS Code, JetBrains, GitHub, GitLab |
| Build |
Compiling code, creating artifacts |
Maven, Gradle, npm, Docker |
| Test |
Automated testing at all levels |
pytest, JUnit, Cypress, Playwright, k6 |
| Security |
Scanning for vulnerabilities |
Snyk, Trivy, Checkov, SonarQube, Semgrep |
| Deploy |
Releasing to environments |
ArgoCD, GitHub Actions, AWS CodeDeploy |
| Operate |
Running and scaling in production |
Kubernetes, Docker, AWS ECS, Terraform |
| Monitor |
Observing system health and performance |
Datadog, Prometheus, Grafana, PagerDuty |
No organization uses all of these tools. The right toolchain depends on your stack, team size, and constraints. Start with the minimum viable pipeline: Git repository, CI/CD system, automated tests, and basic monitoring. Add tools as you encounter specific problems.
DevOps careers in 2026
DevOps skills are among the highest-compensated in the technology industry. Median US salaries (Levels.fyi, Q1 2026):
| Role |
Median salary (USD) |
Key skills |
| DevOps Engineer |
$155,000 |
CI/CD, Terraform, Kubernetes, scripting |
| Site Reliability Engineer (SRE) |
$170,000 |
Monitoring, incident response, SLOs, distributed systems |
| Platform Engineer |
$165,000 |
Internal developer platforms, Kubernetes, IaC |
| DevSecOps Engineer |
$165,000 |
Security scanning, compliance automation, policy-as-code |
| Release Engineer |
$140,000 |
CI/CD pipelines, artifact management, deployment strategies |
The demand is global. African technology companies, fintech startups (Flutterwave, Paystack, Chipper Cash), and global enterprises with African operations are actively hiring DevOps engineers. Nigeria and Kenya are the two largest DevOps job markets on the continent, followed by South Africa, Egypt, and Ghana.
How to start learning DevOps
-
Learn Linux fundamentals. DevOps is built on Linux. You need to be comfortable with the command line, file systems, networking, processes, and package management.
-
Learn Git and GitHub. Version control is the foundation of every DevOps practice. Understand branching strategies, pull requests, merge conflicts, and rebasing.
-
Build a CI/CD pipeline. Pick GitHub Actions (most accessible), create a simple pipeline that builds and tests a web application on every push. This single exercise teaches more than reading ten articles about CI/CD.
-
Learn Docker. Containerize an application. Understand images, containers, Dockerfiles, multi-stage builds, and Docker Compose.
-
Learn Kubernetes basics. Deploy your containerized application to a local Kubernetes cluster (minikube or kind). Understand pods, deployments, services, and ingress.
-
Learn Terraform. Provision cloud resources through code. Deploy a VPC, subnet, and compute instance. Destroy it and rebuild it.
-
Get certified. The AWS DevOps Engineer Professional certification is the gold standard. For entry level, start with AWS Cloud Practitioner or Azure AZ-900, then add DevOps-specific credentials. Citadel Cloud Management's free DevOps course track covers all of these fundamentals.
Frequently asked questions
Is DevOps a job title or a methodology?
Both. DevOps started as a methodology -- a set of practices and cultural principles. Over time, the industry created specific DevOps Engineer, SRE, and Platform Engineer roles to own and implement those practices. The methodology is more important than the title: an organization can adopt DevOps practices without hiring anyone with "DevOps" in their job title.
How is DevOps different from Agile?
Agile focuses on software development -- how teams plan, build, and iterate on software features. DevOps extends that focus to include deployment, operations, and monitoring. Agile answers "how do we build the right software?" DevOps answers "how do we deliver and operate that software reliably?" They complement each other.
Do I need to know how to code to work in DevOps?
Yes, but you do not need to be a software developer. DevOps engineers write scripts (Bash, Python), configuration files (YAML, HCL, JSON), and automation pipelines. You need to read and understand application code to troubleshoot issues, but you are not typically writing application features.
What is the difference between DevOps and SRE?
Site Reliability Engineering (SRE) is Google's implementation of DevOps principles with a specific focus on reliability, availability, and performance. SRE introduced concepts like error budgets, SLOs (Service Level Objectives), and toil reduction. DevOps is the broader philosophy; SRE is a concrete, opinionated implementation of that philosophy.
Can small teams practice DevOps?
Absolutely. DevOps is arguably more impactful for small teams because they cannot afford the inefficiency of manual processes. A two-person startup that sets up CI/CD from day one will outship a ten-person team deploying manually every two weeks.
How long does it take to learn DevOps?
A focused learner can build foundational DevOps skills in 3 to 6 months: Linux, Git, Docker, basic CI/CD, and one cloud provider. Becoming proficient enough for a DevOps Engineer role typically takes 6 to 12 months of dedicated study and hands-on practice. Mastery -- designing resilient systems, managing complex Kubernetes clusters, implementing security at scale -- takes 2 to 3 years of professional experience.
Sources cited
-
DORA State of DevOps Report 2024, Google Cloud DORA, 2024.
-
Fortune Business Insights DevOps Market Report, Fortune Business Insights, 2025.
-
McKinsey Developer Productivity Report, McKinsey, 2024.
-
GitLab DevSecOps Survey 2025, GitLab, 2025.
-
Gartner Platform Engineering Prediction, Gartner, 2024.
-
Accelerate: The Science of Lean Software and DevOps, Nicole Forsgren, Jez Humble, Gene Kim, IT Revolution Press, 2018.
-
The Phoenix Project, Gene Kim, Kevin Behr, George Spafford, IT Revolution Press, 2013.
-
Site Reliability Engineering: How Google Runs Production Systems, Google, 2016.
*Published by Citadel Cloud Management. Start your DevOps journey with our 17 free cloud courses covering DevOps, CI/CD, Terraform, Kubernetes, and cloud security.*
