Work Methodologies DefinedTerm

DevOps

Also known as: DevOps, Dev Ops, Development Operations

A set of cultural practices, processes, and tools that integrate software development and IT operations to accelerate delivery and improve reliability.

Updated: 2026-01-04

Definition

DevOps is a cultural movement and set of practices that unify software development (Dev) and IT operations (Ops) to reduce delivery cycles, increase deployment frequency, and improve system reliability. The term was coined in 2009 by Patrick Debois and Andrew Shafer.

DevOps extends agile principles beyond the development team, including the entire software lifecycle: from planning to deployment, monitoring, and feedback. The goal is to create continuous value flow from code to production.

Core Principles

The Three Ways (from DevOps Handbook):

  1. Flow: optimize workflow from left to right (dev → ops → customer). Minimize batch size, reduce handoffs, automate toil.

  2. Feedback: create rapid feedback loops from right to left. Monitoring, alerting, quick rollbacks. “Fail fast, learn faster”.

  3. Continuous Learning: culture of experimentation and continuous improvement. Blameless postmortems, time for learning, knowledge sharing.

CALMS framework:

  • Culture: collaboration, trust, shared responsibility
  • Automation: CI/CD, infrastructure as code, automated testing
  • Lean: eliminate waste, continuous flow, value-based metrics
  • Measurement: data-driven decisions, visibility on key metrics
  • Sharing: knowledge sharing, incident learning, inner source

Key Practices

Continuous Integration/Continuous Delivery (CI/CD): frequent code integration (multiple times daily) with automated build, test, and deployment. Every commit goes through a pipeline verifying quality and deployability.

Infrastructure as Code (IaC): infrastructure management through versioned code (Terraform, CloudFormation, Ansible). Enables reproducibility, disaster recovery, environment parity.

Monitoring and Observability: metrics, logs, distributed tracing. Tools like Prometheus, Grafana, Datadog, OpenTelemetry. Goal: understand system behavior in production, not just “is it up?”.

Deployment strategies: blue/green deployments, canary releases, feature flags. Reduce risk by enabling quick rollbacks and gradual production testing.

Site Reliability Engineering (SRE): Google’s approach applying software engineering principles to operations. Error budgets, SLO/SLI, toil reduction.

DORA Metrics

DORA (DevOps Research and Assessment) identifies 4 key metrics to measure DevOps performance:

  1. Deployment Frequency: how often releases go to production
  2. Lead Time for Changes: time from commit to production deploy
  3. Mean Time to Recovery (MTTR): time to restore service after incident
  4. Change Failure Rate: percentage of deploys causing production failure

“Elite performers” (State of DevOps 2023):

  • Deploy on-demand (multiple times daily)
  • Lead time under 1 hour
  • MTTR under 1 hour
  • Change failure rate under 5%

Practical Considerations

Cultural transformation: DevOps requires deep cultural change. Common resistance in organizations with established silos, rigid processes, blame culture. Leadership buy-in is critical.

Toolchain: the market offers hundreds of tools (Jenkins, GitLab CI, GitHub Actions, ArgoCD, Kubernetes, etc.). Risk is “tool-driven DevOps” ignoring culture. Principle: people over process over tools.

Security shift-left: DevSecOps integrates security from design. Automated SAST/DAST in pipeline, container vulnerability scanning, compliance as code.

Platform Engineering: recent evolution creating self-service platforms for developers. Reduces cognitive load and standardizes best practices. Platform teams manage IDP (Internal Developer Platform).

Use Cases and Impact

Innovation velocity: Netflix deploys thousands of times daily. Amazon every 11.7 seconds (2015). This velocity enables rapid experimentation and real user feedback.

Reliability: Google maintains 99.99% uptime with SRE practices. Automation reduces human errors, frequent deployments reduce problem blast radius.

Developer productivity: eliminating toil and wait time (e.g., waiting for Ops for provisioning) increases flow and developer satisfaction. Direct correlation with retention.

Common Misconceptions

”DevOps is a separate role/team”

No. “DevOps teams” often recreate silos. DevOps is culture: dev and ops work together, share on-call, use same tools. Platform teams are acceptable if enablers, not gatekeepers.

”DevOps means firing Ops”

False. It transforms the role: from ticket-driven manual work to platform engineering, automation, reliability engineering. Ops skills are more critical, not less.

”DevOps is only for startups/web companies”

No. Regulated enterprise organizations (banking, healthcare) successfully adopt DevOps. Requires adaptations (e.g., compliance automation) but principles apply universally.

”You must adopt everything at once”

No. DevOps is an incremental journey. Start small: pick one bottleneck, apply one practice (e.g., automated testing), measure impact, iterate. Big-bang transformations fail.

Sources

Back to Glossary