Definitions

DevOps — integration of software development (Dev) and IT operations (Ops) to shorten systems development lifecycle and provide continuous delivery of software.

In simple words, DevOps is a bridge between code and «healthy run in production» through automation.

1 2 3 4

___________Dev____________ / \ code -> build -> test -> release -> deploy -> operate -> monitor \_______________Ops_______________/

CI/CD pipelines:

• Continuous Integration (build, test, merge)
• Continuous Delivery (to repo) and Deployment (to production)

Related terms:

• Software Architecture at Scale (Systems Design)
• DevSecOps — also takes Security into account

Modern evolution of DevOps:

• SRE (Site Reliability Engineering) — reliability, scalability, automation, recovery
• Platform engineering — IDP (Internal Developer Portals), DX, CI/CD architecture, infrastructure abstraction
• Infrastructure engineering — focus on underlying cloud resources (usually AWS and GCP — Amazon and Google)

Software

• GitHub Action, GitLab, Jenkins — CI/CD
• Nix/NixOS — reproducible builds
• TerraForm — IaC (Infrastructure as Code) for cloud infrastructure
• Docker — containerization
• Kubernetes — orchestrator of containers
• Prometheus, Graphana — monitoring, alerting, visualizing data

DORA Metrics

DORA:

• Originally a name of research startup [D]ev[O]ps [R]esearch and [A]ssessment
• Industry then widely adopted DORA metrics that this startup popularized

DORA metrics:

1. Deployment Frequency — how often do we ship
2. Lead Time to Change — how long from git commit to production
3. Change Failure Rate — how often do we break production
4. MTTR ([M]ean [T]ime [T]o [R]estore service) — how fast we fix when smth breaks