Exam AWS Certified Machine Learning - Specialty topic 1 question 298 discussion

https://docs.aws.amazon.com/sagemaker/latest/dg/shadow-tests.html

Shadow deployment is new technique provided by sagemaker

Answer: C Explanation: The best solution for this scenario is to use shadow deployment, which is a technique that allows the company to run the new experimental model in parallel with the existing model, without exposing it to the end users. In shadow deployment, the company can route the same user requests to both models, but only return the responses from the existing model to the users. The responses from the new experimental model are logged and analyzed for quality and performance metrics, such as accuracy, latency, and resource consumption12. This way, the company can validate the new experimental model in a production environment, without affecting the current live traffic or user experience

Shadow deployment consists of releasing version B alongside version A, fork version A’s incoming requests, and send them to version B without impacting production traffic. This is particularly useful to test production load on a new feature and measure model performance on a new version without impacting current live traffic. source: https://aws.amazon.com/blogs/machine-learning/deploy-shadow-ml-models-in-amazon-sagemaker/