Kubernetes helps you reduce the risk of errors and shorten the time it takes for containerized apps to enter the market. It is, however, a significant undertaking and a sizable endeavor.
We have developed CNO Deploy to help you implement the best continuous deployment model that suits your business needs.
CNO Deploy brings five strategies to help you deliver faster applications.
CNO has a component called the CNOCD operator. It continuously monitors the image registers of the applications deployed on your clusters. So, it can detect new image publication events and list the latest versions of an application.
This way enables you to either do continuous delivery or continuous deployment.
A Continuous Delivery model refers to a manual method. You can display the latest versions of your application and choose one of them to deploy using command line cnoctl or the CNO-UI.
You can also choose, by selecting "Auto Deploy", a Continuous Deployment model.
A Continuous Deployment model is an automatic way to deploy your applications. The operator monitors the image registry of your application, and whenever you push a new image, it detects it and deploys it automatically.
Advanced Deployment Strategies¶
Blue / Green¶
This deployment model releases two versions of the application simultaneously in production environments:
The first, Blue, is unchanged and is automatically deployed on the market.
The second, Green, is available for review by internal teams but remains offline.
It is up to you to choose the right moment to release the Green (new) version with the freedom to roll back to the Blue (old) Version.
Canary is quite similar to the Blue/Green strategy.
You can deploy your app in two versions directly in the production environment:
The new one will be accessible and tested by a defined consumer sample;
The old one will remain in production for most consumers.
The transition from each other is staggered: 10 %, 20 %, up to 100% of the new app with the freedom to roll back if the new version is not viable.
A/B testing is the best way to improve users' experience by testing two versions of your app on consumer groups A & B.
Unlike Canary and Blue/Green counterparts, A/B testing is a behavioral monitoring paradigm and not an engineering process.
Under the behavioral spectrum, KPIs received through A/B testing will guide I.T. & marketing teams to make the best decision.