Continuous Integration–Continuous Delivery (CI-CD)

---

Continuous Integration–Continuous Delivery (CI-CD) pipeline is an automated DevOps workflow that streamlines the software delivery process.

• key aspect is automation to ensure code quality
• automation performs quality control
  • assesses performance, APIs, and security

CI/CD Pipeline vs CI/CD

• CI/CD pipeline is the agile DevOps workflow
• CI/CD is the combined practice of continuous integration and continuous delivery
• CI/CD streamlines:
  • application code
  • testing
  • development processes

Benefits of CI/CD Pipeline

• allows DevOps teams to write code, integrate it, run tests, deliver releases, and deploy changes to software collaboratively and in real-time
• Automation of software releases
• Reduced deployment time
  • automated testing
  • continuous deployment and automated provisioning allow changes to go live in minutes
• Decreased cost
  • faster development means less time spent
  • results in reduced cost
• Continuous feedback loop
  • continuous cycle
    • build, test, deploy
  • enable devs to act quickly
• Early error detection
  • in continuous integration, testing is automated to look for issues in integration
• Reduced downtime and enhanced reliability
  • automated rollbacks for deployment failures, bugs, performance issues
• Improved team collaboration and system integration
  • all team members can change code, respond to feedback and issues
  • single repo

How it Works

• CI/CD pipeline consists of 3 connected methodologies:
  • Continuous integration
  • Continuous delivery
  • Continuous deployment

Continuous Integration

Continuous integration (CI) is a software development practice in which developers regularly integrate new code and code changes into a central code repository throughout the development cycle.

• central code repo is managed by a version control system (VCS)
• allows devs to work independently
  • create own branch to implement changes
• individual work is then pushed to an automated system
  • uses scripts to build and test the code changes
• CI server then compiles source code changes into the main branch
• enables frequent submission of code changes

Continuous Delivery

Continuous delivery (CD) automates the progress of software code changes through testing and eventual release into production.

• puts validated code changes into select environments or code repositories
• code moves through multiple production-like environments (QA, staging, etc.) for testing
• software and APIs are tested and errors resolved through an automated process
• goal is to deploy new code with minimal effort, but allow for human oversight

Continuous Deployment

Continuous deployment (CD) is a strategy in software development where code changes to an application are released automatically into the production environment.

• once code changes are approved from delivery, it is deployed to production via automation
  • no human intervention

Machine Learning Pipeline

A machine learning pipeline is a series of interconnected data processing and modeling steps designed to automate, standardize and streamline the process of building, training, evaluating and deploying machine learning models.

• integrate with DevOps practices to enable CI/CD of machine learning models
  • called machine learning operations (MLOps)

Stages of CI/CD

• Build
  • involves creating and compiling code
  • teams build off source code collaboratively and integrate new code quickly to resolve conflicts
• Test
  • automated tests occur
    • integration test, unit test, regression tests
• Deliver
  • approved codebase is sent to a production environment
    • only automated after dev approval
  • code is sent to repositories then moved into production or deployment by human
• Deploy
  • changes are deployed to production
  • unlike continuous delivery, this is fully automated

CI/CD Security

CI/CD security focuses on practices, processes, and technologies that implement and manage security and compliance measures across the CI/CD pipeline.

• DevSecOps automates integrating security practices into the pipeline
  • shift-left testing
    • involves integrating security testing earlier in the SDLC
  • shift-right testing
    • involves testing later in development, usually in prod
  • spreads testing across SDLC

Tools

• Jenkins
  • automated CI server written in Java
• Travis CI
• CircleCI
• All major cloud providers offer cloud-based CI/CD toolchains
• In cloud environments, applications are packaged and shipped in containers
  • Docker for containers
  • Kubernetes for container orchestration