All at once

• Deploys to all instances simultaneously.
• You will experience a total outage.
• Not ideal for mission-critical production system.
• Rolling back: another outage to get back to the previous version

Rolling Deployment Policy

• Deploys the new version in batches
• Each batch is taken out of service while the deployment takes place
• You env capacity will be reduced by the number of instances in a batch while the deployment takes places.
• Not ideal for performance sensitive system.
• Rolling back: perform an additional rolling update to roll back the changes

 

Rolling with Additional Batch

• Launches an additional batch of instances.
• Deploys the new version in batches
• Maintains full capacity throughout the deployment
• Rolling back: perform an additional rolling update to roll back the changes

 

Immutable Deployments

• Deploy the new version to a fresh group of instances
• Only when the new instances pass their health checks, should the old instances be terminated.
• This is the preferred approach for mission cirtical system.
• Rollback: Just delete the new instances