How to Deploy a New Vertex AI Model Version with Minimal Disruption?

You are deploying a new version of a model to a production Vertex Al endpoint that is serving traffic. You plan to direct all user traffic to the new model. You need to deploy the model with minimal disruption to your application. What should you do?

1. Create a new endpoint
1. Create a new endpoint
1. Create a new model. Set the parentModel parameter to the model ID of the currently deployed model. Upload the model to Vertex AI Model Registry. Source Reference Answer
1. Create a new model. Set it as the default version. Upload the model to Vertex AI Model Registry

Community Votes

100%

100% of anonymous learners picked answer C. Votes are pick records left by other test-takers — they are not the verified answer.

Community Insight

Tests understanding of Vertex AI model versioning and endpoint management, with the common trap being the assumption that creating a new endpoint or simply marking a model as default automatically routes traffic without configuration.

Deploying a new model version to an existing Vertex AI endpoint using the parentModel parameter ensures seamless traffic routing and avoids service interruptions. The community strongly agrees that leveraging the existing endpoint URL and model registry versioning minimizes application disruption during updates.

Option D is frequently chosen by candidates who assume setting a model as the default version instantly shifts traffic, overlooking that explicit endpoint deployment and traffic allocation settings are required to avoid routing to the legacy version.

Community Discussion (5 comments)

fitri001 👍 3 Selected: C

Minimal Downtime: By deploying the new model to the existing endpoint, you avoid any service interruptions caused by creating and switching to a completely new endpoint. Versioning: Setting the parentModel parameter allows you to track the lineage of your models and easily revert to the previous version if needed. Traffic Control: Vertex AI lets you control traffic allocation between different versions of a model deployed on the same endpoint. Setting the new model to 100% traffic directs all user requests to the new version.

guilhermebutzke 👍 2 Selected: C

My Answer: C In the context of deploying machine learning models, setting the parentModel parameter to the model ID of the currently deployed model means that the new model being deployed is created as a child model or an iteration of the existing model. This allows the new model to inherit certain properties or characteristics from the existing model, such as the architecture, hyperparameters, or feature transformations. Create a new Endpoint is Unnecessary.

sonicclasps 👍 1 Selected: D

Optionally set this model as the default version. The default version is preselected whenever the model is used for prediction (although you can still select other versions). https://cloud.google.com/vertex-ai/docs/model-registry/versioning

BlehMaks 👍 1 Selected: C

a,c -creating new endpoint is an unnecessary disruption to the application d - doesn't work, two models are on the same endpoint and traffic is still going through the old model

pikachu007 👍 3 Selected: C

Leverages existing endpoint: Using the same endpoint maintains the same endpoint URL, avoiding DNS updates and potential service interruptions. Gradual traffic transition: Vertex AI allows you to gradually shift traffic between model versions, ensuring a smooth transition without impacting users. Clear versioning: Setting parentModel establishes a relationship between the new model and the existing one, aiding in organization and tracking model lineage.

Comments & Corrections

No comments yet — spotted an error or have a note? Share it below.

Expert Analysis

Why the Answer Is Correct

Creating a new model version via the Model Registry while specifying the parentModel parameter establishes a clear lineage between the old and new iterations. This approach allows you to deploy the updated artifact to the existing production endpoint, preserving the original endpoint URL and preventing client-side DNS or connection breaks. Vertex AI natively supports gradual traffic shifting between versions on the same endpoint, enabling zero-downtime deployments.

Why the Other Options Are Wrong

Options A and B suggest creating a completely new endpoint, which forces clients to update their endpoints and introduces unnecessary network propagation delays and potential downtime. Option D incorrectly assumes that merely marking a model as the default version automatically redirects live traffic; without explicitly deploying the version to the endpoint and configuring traffic weights, requests will continue routing to the currently active version.

Community Comment Notes

Multiple high-voted comments emphasize that reusing the existing endpoint maintains the same URL, eliminating DNS updates and service interruptions [1][2]. Contributors note that the parentModel parameter enables proper lineage tracking and easy rollback capabilities [3]. Others clarify that simply setting a default version does not override active endpoint routing configurations, making D ineffective for immediate traffic migration [5].

Official Reference

Exam Strategy

Always prioritize solutions that reuse existing infrastructure components like endpoints to minimize client-side changes and latency. When managing model updates, focus on versioning controls and traffic allocation mechanisms rather than recreating resources, as this aligns with cloud-native MLOps best practices for zero-downtime deployments.

Related Analysis

Practice All PMLE Questions

Access 65 questions with complete answers and detailed explanations.

View Full PMLE Practice Test →

← Back to PMLE Study Guide