Canary Deployment
The short answer · 23 words
In short
Deployment strategy in which a new software version is initially rolled out and observed only for a small proportion of users or traffic.
Flow-alerta.cab (2026): Canary Deployment. https://flow-alerta.cab/en/glossar/canary-deployment. Editorially reviewed on: 09 October 2026.
How much detail would you like?
Practical
Can limit the risk of large releases and is particularly suitable for web applications, APIs and AI services, whose behavior should be monitored under real traffic.
Example scenario: A new RAG version initially receives five percent of productive traffic; error rate, response quality and latency are compared with the previous version.
Set up automated metrics, defined rollout levels and fast rollback for high-risk releases; consider data migrations separately.
At a glance
- Core
- In a Canary deployment, old and new versions run in parallel. Only a small part of the traffic initially reaches the new version. If defined quality and error metrics are adhered to, the proportion is gradually increased; in the event of problems, the rollout can be stopped or reset.
- Use
- Can limit the risk of large releases and is particularly suitable for web applications, APIs and AI services, whose behavior should be monitored under real traffic.
- Example
- Example scenario: A new RAG version initially receives five percent of productive traffic; error rate, response quality and latency are compared with the previous version.
- Limits
- Roll out without clear cancellation criteria; do not make new and old versions cleanly observable; do not plan database changes backwards compatible; equate Canary with A/B test.
Questions about this term
Is Canary Deployment the Same as Blue-Green?
No.
Why only part of the traffic?
To limit the effect of errors.
Do we need monitoring?
Yes.
All details and background
In a Canary deployment, old and new versions run in parallel. Only a small part of the traffic initially reaches the new version. If defined quality and error metrics are adhered to, the proportion is gradually increased; in the event of problems, the rollout can be stopped or reset.
ELI5 – Explained simply
A new version is first delivered to only a few users. If it works stable, they all gradually get.
Why it matters in practice
Can limit the risk of large releases and is particularly suitable for web applications, APIs and AI services, whose behavior should be monitored under real traffic.
Example
Example scenario: A new RAG version initially receives five percent of productive traffic; error rate, response quality and latency are compared with the previous version.
Pitfalls and limits
- Roll out without clear cancellation criteria; do not make new and old versions cleanly observable; do not plan database changes backwards compatible; equate Canary with A/B test.
Origin & history
Google Cloud Deploy documents canary rollouts that gradually increase the traffic share of a new version.
Use in mid-sized companies
- Set up automated metrics, defined rollout levels and fast rollback for high-risk releases; consider data migrations separately.
Related terms
- Rollback
Matching Flow-alerta service
Websites & web applicationsWebsites & web applications: technical implementation and testing of a suitable web application.Matching tool / use case
View use casesTool example: Google Cloud Deploy; Use Case: step-by-step rollout of a new web or AI service.Sources
- docs.cloud.google.comGoogle Cloud, Canary deployment strategy
Quellenprüfung: 09.10.2026. Definition und dokumentierte Einordnung abgeglichen; Beispiele beschreiben mögliche Anwendungen.
Editorially reviewed on: