GPU quota increase not reflecting in compute engine

[sarath7974]

Increased quota for both Committed NVIDIA A100 80GB GPUs in us-east4  and Committed NVIDIA A100  GPUs in asia-southeast1. And 1 quota was allocated for each , but when i tried creating vm with gpu and region its still saying quota not available.

error showing when creating vm :

[almirajean]

Hi @sarath7974,

Welcome to Google Cloud Community!

Based on the screenshot provided you request a quota in Committed NVIDIA A100 80GB GPUs instead of NVIDIA A100 80GB GPUs. Committed is a quota of GPU that you can commit (reserve) in your project while NVIDIA A100 GPUs is a quota of GPU that you can use in your project (attached to a GCE). 

In addition, when you request a GPU quota  you must request a quota for the GPU models that you want to create in each region and an additional global quota (GPUs (all regions)) for the total number of GPUs of all types in all regions.

Here’s a guide for requesting quota increase : 

1. Go to "IAM & Admin” > “Quotas & System Limits”.
2. On the “Filter” section select “Name: NVIDIA A100 80GB GPUs” choose the one without Committed.

3. Find the location us-east4 or the location you want to increase the quota. Then click “Edit quota” you will prompt in to enter the Quota changes page. 

4. You must also request a quota for the GPU models in Global quota GPUs (all regions).
5. You will receive a response from the Compute Engine team within 24 to 48 hours of the status of your request.

For additional information you may check this Requesting Quota documentation.

I hope the above information is helpful.