Default VM instance type does not provide enough cores

[yksnilowyrahcaz]

Please provide us with the following information:

This issue is for a: (mark with an x)

- [x] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

Minimal steps to reproduce

Run python -m deployment.deploy

Any log messages given by the failure

azure.core.exceptions.HttpResponseError: (BadRequest) The request is invalid.
Code: BadRequest
Message: The request is invalid.
Exception Details: (InferencingClientCreateDeploymentFailed) InferencingClient HttpRequest error, error detail: {"errors":{"VmSize":["Not enough quota available for Standard_DS3_v2 in SubscriptionId [REDACTED]. Current usage/limit: 0/6. Additional needed: 8 Please see troubleshooting guide, available here: https://aka.ms/oe-tsg#error-outofquota"]},"type":"https://tools.ietf.org/html/rfc7231#section-6.5.1","title":"One or more validation errors occurred.","status":400,"traceId":"[REDACTED]"}
Code: InferencingClientCreateDeploymentFailed
Message: InferencingClient HttpRequest error, error detail: {"errors":{"VmSize":["Not enough quota available for Standard_DS3_v2 in SubscriptionId [REDACTED]. Current usage/limit: 0/6. Additional needed: 8 Please see troubleshooting guide, available here: https://aka.ms/oe-tsg#error-outofquota"]},"type":"https://tools.ietf.org/html/rfc7231#section-6.5.1","title":"One or more validation errors occurred.","status":400,"traceId":"[REDACTED]"}

Expected/desired behavior

VM instance type that provides minimum number of cores for example to run.

OS and Version?

Windows 10

Versions

Cloned this example from commit 065ef02

Mention any other details that might be useful

It appears that Standard_DS3_v2 is the VM instance type in deploy.py. This VM instance provides 4 cores. Per the traceback, it seems that 8 cores are required for this example. Does the VM instance type need to be something that provides at least 8 cores?

Thank you in advance for your consideration of this inquiry.

---

Thanks! We'll be in touch soon.

[dudimasta]

You need to request for more resources in Quotas for region you are deploying, provider type "Machine Learning" (not compute).

On default quotas (without requesting for more resources) minimal machine learning quotas should work. Try to change in deploy.py line 61, e,g.:
instance_type="Standard_F2s_v2"
and rerun
python -m deployment.deploy --endpoint-name <...> --deployment-name <...>
(valid sizes: https://learn.microsoft.com/en-us/azure/machine-learning/reference-managed-online-endpoints-vm-sku-list?view=azureml-api-2)