Share via

Is the "Enrich a search index using Azure Machine Learning model" MS Learn exercise impossible to do with a Student Azure Subscription?

Jacob Oros 0 Reputation points
2024-09-22T20:50:19.97+00:00

In the "Create a regression training pipeline" section of the lab, you are required to "Create an inference cluster for the endpoint" so that you will be able to connect the Search AI index to the custom skill at that endpoint.
When you create the Kubernetes cluster (AksCompute), any VM size you select, (including the Standard_A2_v2 that the lab suggests) will fail to deploy if you're using an Azure for Students subscription. I get the error:
Status Code: 400 BadRequest Reason Phrase: Bad Request Response Body: { "code": "ErrCode_InsufficientVCPUQuota", "details": null, "message": "Insufficient vcpu quota requested 6, remaining 4 for family standardAv2Family for region uksouth.", "subcode": "" }
Even though the VM they suggest using says it should only take up 2 cores, the request is for 6, no matter what I tried. Copilot suggested it might be because of deploying multiple nodes, but I do not know how to check if that is the case. (It is my first encounter using kubernetes)

The Student Azure subscription has a maximum quota of 4 Standard Av2 Family vCPUs and it cannot be increased. Am I doing something wrong trying to deploy the kubernetes cluster or there is just no way to complete this lab with an Azure Student Subscription.

Azure Machine Learning
Azure Machine Learning
An Azure machine learning service for building and deploying models.
2,887 questions
Azure AI Search
Azure AI Search
An Azure search service with built-in artificial intelligence capabilities that enrich information to help identify and explore relevant content at scale.
992 questions

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. romungi-MSFT 45,721 Reputation points Microsoft Employee
    2024-09-23T07:48:26.2133333+00:00

    @Jacob Oros Student subscription quotas come with limited compute options and quotas that could hamper creating certain compute types. If you have chosen a certain compute type that is within your limits but if you still see an error, then it means that at the subscription level there might some other resource or compute that is running that is using up the quota. Try to delete any unused compute or endpoints that are running and try to check deploying after the delete.

    As per my opinion, using a personal subscription under free tier limits might help for basic scenarios if you are just trying to complete a basic course. Thanks!!

    If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.

    1 person found this answer helpful.
    0 comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.