Title: Error 429: Exceeding Quota Limit When Creating a Model Version in AI Platform

[prosperwashaya]

I am attempting to create a new version of a model in AI Platform using gcloud commands, but I am encountering a quota error related to the n1-standard-2 machine type.

My goal is to deploy a new version of my model, to AI Platform to leverage GCP's scalable infrastructure for serving predictions. As follows:

!gcloud ai-platform models create {MODEL_NAME} \
--project {PROJECT} \
--region {REGION}

!gcloud ai-platform versions create {VERSION_NAME} \
--project {PROJECT} \
--region {REGION} \
--model {MODEL_NAME} \
--origin {EEIFIED_DIR} \
--framework "TENSORFLOW" \
--runtime-version=2.3 \
--python-version=3.7\

When attempting to create the model version, I received the following error:

"error": {
"code": 429,
"message": "The requested number of n1-highmem-2 exceeds the quota limit. Current usage/limit: 0/0, Requested: 8.",
"status": "RESOURCE_EXHAUSTED"

What could be the issue?

[Poala_Tenorio · Accepted solution]

The error message you're seeing indicates that you've exceeded the quota limit for the "n1-standard-2" machine type when attempting to create a new version of your model in AI Platform. The message specifies that you've requested 8 instances of the "n1-standard-2" machine type, and the current usage/limit is 0/0, which suggests that the quota limit for this machine type has not been set or is set to zero.

To resolve this issue, you should follow these steps:

First, you should verify the quota limits for the "n1-standard-2" machine type in your Google Cloud project. You can do this through the Google Cloud Console.

1. Go to the Google Cloud Console.
2. Click on the navigation menu and select "IAM & Admin" > "Quotas."
3. In the filter box, type "n1-standard-2" to locate the relevant quota.
4. Check the "Limit" value for the "n1-standard-2" machine type. If it's set to zero or a low value, that's the cause of the issue.

If the quota limit for "n1-standard-2" is set to a value of 0 or insufficient for your needs, you should request a quota increase.

1. In the Quotas page in the Google Cloud Console, click on the "Edit Quotas" button for the "n1-standard-2" quota.
2. Fill out the quota increase request form, providing the rationale for your request and the number of instances you need.
3. Submit the request, and Google Cloud support will review it.

You can also consider using a different machine type that is available within your existing quota limits. You can specify a different machine type when creating your model version. For example, you can use "n1-standard-4" or another machine type with higher quotas.