Guide-linked quiz · Medium Level
Blue-Green Deployment for ML Models Quiz
Understand traffic cutover and rollback in blue-green releases, including capacity, schema compatibility and limits of routing rollback.
Question 1 of 8
How are blue and green environments assigned during a release?
Preview every question8 questions
Answers and explanations appear as you take the quiz.
- How are blue and green environments assigned during a release?
- One serves current traffic while the other stages a candidate release
- They store separate training and test labels
- They run different model classes permanently
- They split the GPU into two hardware drivers
- What enables a fast routing rollback after a bad cutover?
- The new model is retrained during every request
- The previous environment remains available and healthy
- The database schema is deleted
- The old artifact was overwritten
- Why plan capacity for blue and green simultaneously?
- The model runs on both environments after cleanup forever
- Traffic is always doubled by the router
- Both production-like environments may need resources during rollout
- Blue-green requires two datasets
- Why should database changes remain compatible with both versions during cutover?
- Green must rewrite blue's weights
- The router changes database columns automatically
- ML models do not read data schemas
- A rollback may return traffic to the older application or model
- What does shadow traffic provide?
- An automatic full production switch
- Candidate predictions for comparison without serving them to users
- A guarantee of zero latency
- A replacement for model evaluation data
- What does routing rollback not necessarily undo?
- The router's traffic target
- Which environment receives future requests
- External side effects or writes already made
- The old environment's availability
- How does a canary relate to the basic blue-green cutover?
- It only validates container image signatures
- It is another name for two environments with no traffic shift
- It eliminates the need for a green environment
- It can expose a small traffic fraction before the full switch
- Why warm a large model in the green environment before switching?
- It can reduce initialization delay and accelerator allocation surprises
- It guarantees output quality
- It eliminates the old environment's costs
- It validates every subgroup metric
Keep testing yourself
More quizzes picked for your level.