For Azure Machine Learning online endpoints, a way to send a copy of up to 50% of live requests to a test deployment. Clients still get answers only from the live deployment.
Also called shadow testing.
Read more: Microsoft Learn
In the Ultra Transcenders books
Each book explains Traffic mirroring in context, with comparison tables and the common traps.
Terms in this definition
- Azure Machine Learning
Azure platform where you train, deploy, monitor and retrain models of your own, practising MLOps through tools such as pipelines, online endpoints, model monitoring and prompt flow.
- Online management group
Workloads that may talk directly to or from the internet, or that need no virtual network at all, go in this management group. It is a sibling of Corp and Local beneath Landing zones.
- Model deployment
Inside a resource, each deployment is a named copy of a Foundry model with its own TPM quota and deployment type. Requests identify the model by this deployment name.
- Get
Key Vault permission on secrets that allows a single secret to be read; App Service Key Vault references need nothing beyond it.
Related terms
- Managed online endpoint
Azure Machine Learning online endpoint where Azure provisions and runs the compute for you; it offers autoscale, traffic mirroring and splitting traffic between deployments.
- SPAN session
An older term, still seen in aCLI and AHV alerts, for what is now Traffic Mirroring: duplicating packets from a physical interface or VM to a single receiving VM.