08.03 · Concept
Strategies Compared
Choose between rolling, blue-green and canary on rollback speed, cost and blast radius.
No video curated for this lesson yet
This lesson is written, ordered and part of the path - the video slot is the only thing still open. We are working through Deployment lesson by lesson; 29 of 56 have their video so far.
The written notes below cover this idea in full - you lose nothing by reading instead of watching.
Rolling, blue-green and canary releases trade cost, rollback speed and user impact differently. Rolling is cheap but may run mixed versions. Blue-green gives a fast traffic switch but needs duplicate capacity. Canary limits early exposure, but only works when routing and metrics can detect bad behaviour quickly.
What this lesson answers
- rolling vs blue green vs canary deployment
- which deployment strategy has fastest rollback
- how to reduce blast radius during releases
Notes
Release Strategies Compared — Release strategies exist to change production software while limiting outage risk; without choosing the right strategy, a bad deploy can affect 100% of users, require slow rollback, or double infrastructure cost unexpectedly.
Key Concepts: - Rolling deploy: replace instances gradually, e.g. Kubernetes Deployment with `maxUnavailable: 1` and `maxSurge: 1` updates pods one or two at a time instead of all at once. - Blue-green deploy: run two full environments, `blue` and `green`, then switch traffic 100% at once; cost is approximately …
References
Common questions
- When should I use a rolling deployment?
- Use rolling deployment for routine service changes where extra infrastructure cost matters and gradual instance replacement is acceptable. It is a good default for many stateless services, especially under orchestrators such as Kubernetes. The main risk is mixed-version behaviour, so API and data changes must remain compatible during the rollout.
- Why is blue-green rollback usually faster?
- Blue-green keeps the old and new environments available at the same time, then moves production traffic between them. If the new version misbehaves, rollback can be a traffic switch back to the previous environment rather than rebuilding instances. The tradeoff is paying for near-duplicate serving capacity during the overlap.
- What makes canary safer than a normal release?
- Canary releases send only a small share of traffic to the new version first, so a bad change affects fewer users while metrics are checked. They are best for risky changes where limiting blast radius matters. They depend on reliable routing, meaningful health signals and enough traffic to expose failures.
Short definition: what is Strategies Compared?