CompTIA Cloud+ (CV0-004)DeploymentHard
A cloud operations team is preparing to deploy a critical new application into production. Before the full production rollout, they want to perform a deployment to a small subset of real users to gather feedback and monitor for issues, while the majority of users continue to use the old version. If issues arise, they want to quickly revert to the old version without impacting the main user base. Which deployment strategy does this describe?
- ABig Bang Deployment
- BBlue/Green Deployment
- CCanary Deployment
- DRolling Deployment
Show answer & explanationAnswer & explanation
Correct answer: C. Canary Deployment
Canary deployment involves gradually rolling out a new version of an application to a small subset of users. This allows for testing in a real production environment, monitoring for issues, and quickly reverting if necessary, minimizing risk before a broader rollout. This matches the scenario perfectly.
Why the other options are wrong
- A. Big Bang deployment releases the new version to all users simultaneously, which is high risk and does not allow for phased testing or easy rollback to the old version.
- B. Blue/Green deployment runs two identical production environments (blue for old, green for new) and switches traffic wholesale, allowing for quick rollback but typically a larger initial user base for the new version.
- D. Rolling deployment replaces instances of the old version with the new version gradually, but once an instance is updated, it serves new code, making quick, full rollback more complex than canary.
Canary Deployment
A deployment strategy where a new version of an application is incrementally rolled out to a small subset of users, allowing for real-world testing and easy rollback.
- Reduces risk by exposing new code to a small audience first.
- Allows monitoring performance and errors in production.
- Enables quick rollback if critical issues are found.
Memory trick: Canary tests a few, Blue/Green switches all, Rolling updates gradually, Big Bang goes all at once.