Google Compute Engine can't find GPU even though quota exists

Question

I requested and received my quota for a preemptible K80 GPU.

In the Console I can see this quota available (in region 'us-east1'). However, my VM in the same region is unable to find this GPU:

tensorflow-1-vm: {"ResourceType":"compute.v1.instance","ResourceErrorCode":"404","ResourceErrorMessage":{"code":404,"errors":[{"domain":"global","message":"The resource 'projects/dlp-project-212017/zones/us-east1-b/acceleratorTypes/nvidia-tesla-k80' was not found","reason":"notFound"}],"message":"The resource 'projects/dlp-project-212017/zones/us-east1-b/acceleratorTypes/nvidia-tesla-k80' was not found","statusMessage":"Not Found","requestPath":"https://www.googleapis.com/compute/v1/projects/dlp-project-212017/zones/us-east1-b/instances","httpMethod":"POST"}}

When I use the command-line to check if I have quota, I also see that I do not have available quota.

Anyone knows how to handle these discrepancies? Mainly, how do I run my VM with my GPU now?

Answer 1

The nvidia-tesla-k80 GPu is not available on the zone "us-east1-b" it just available on these zones :

us-west1-b
us-central1-a
us-central1-c
us-east1-c
us-east1-d
europe-west1-b
europe-west1-d
asia-east1-a
asia-east1-b

You can check this Google Document for the Availability Zone of each type of GPU