Sagemaker GPU Instance quota limits

Question

Hi Team,
Writing to you regrading the exception received in instantiating the ml.g4d.xlarge instance which required for one of usecase, I am currently has 12 months free service aws account , when I tried to launch to notebook instance it throws following error.
"
ResourceLimitExceeded
The account-level service limit 'ml.g5.xlarge for notebook instance usage' is 0 Instances, with current utilization of 0 Instances and a request delta of 1 Instances. Please use AWS Service Quotas to request an increase for this quota. If AWS Service Quotas is not available, contact AWS support to request an increase for this quota.
"
But I am able to launch c5,t2 instances successfully, can someone help me out how to get access to this and update limitations.

Answer

Hello. The quota is different for different instance type. You can raise an increase of the quota for Notebook instance ml.g4dn.xlarge or ml.g5.xlarge via Service Quotas.

Upon logging in the AWS Console, you can go here https://us-east-1.console.aws.amazon.com/servicequotas/home/services. Make sure you choose the right region (the above link defaults to N. Virginia region). Then type in "SageMaker" on the "AWS services" search bar on that page and click the only "Amazon SageMaker" in the result. Then find "ml.g5.xlarge for notebook instance usage" and "ml.g4dn.xlarge for notebook instance usage" categories. For each, when you click, it will show "Request quota increase" button. Click that button and start the process of requesting limit increase for those instances.

Sagemaker GPU Instance quota limits

Relevant content