Should I still get a GPU quota increase when the limit is already 1?

1
votes

I am new to Google Cloud Platform and I am struggling to create a VM Instance with a GPU. I believe the first step is to upgrade the account and ask for a quota increase as the free tier does not provide them. I upgraded to the paid account but I already have the limit as 1 for GPUs. I still anyway tried to request an increase from 1 to 1 (which I felt made no sense) and got a mail stating they are unable to grant me additional quota at this time.

enter image description here

But since my limit was already 1, I tried to create an instance anyway and it always shows the following error: "The zone 'projects/project-name/zones/zone-name' does not have enough resources available to fulfill the request. Try a different zone, or try again later."

I tried with different zones but I always seem to get the same error. But, I could create an instance without a GPU. Any help setting up an instance with GPU is appreciated!

1
google-cloud-platformgoogle-compute-engine
This error indicates a shortage of resources in the zone, not a lack of quota. – robsiemb

1 Answers

0
votes

Let's start with the error message you've got:

    The zone 'projects/PROJECT_ID/zones/ZONE_NAME' does not have enough resources available to fulfill the request. Try a different zone, or try again later.

This means that zone ZONE_NAME doesn't have enough resources to proceed with your request at the moment.

The cause of such behaviour: when you stop an instance it releases some resources like vCPU and memory, when you start an instance it requests resources like vCPU and memory back and if there's not enough resources available in the zone you'll get an error message like you've got. More information you can find in the documentation:

This error is due to Compute Engine resource obtainability, and is not due to your Compute Engine quota.

As you can see, this behaviour has no connections to quotas and it's not an issue.

The reason of this behavior is resource availability which depends from users requests and therefore are dynamic.

There are a few ways to solve it:

  1. Move your instance to another zone by following instructions, be sure that new zone contains required GPU and you have enough quota to use it.
  2. Wait for a while and try to start your VM instance again.
  3. Reserve resources for your VM by following documentation to avoid such issue in future:

Create reservations for Virtual Machine (VM) instances in a specific zone, using custom or predefined machine types, with or without additional GPUs or local SSDs, to ensure resources are available for your workloads when you need them. After you create a reservation, you begin paying for the reserved resources immediately, and they remain available for your project to use indefinitely, until the reservation is deleted.