Resource availability for mortals

[stackoverthrow]

I am trying to simply run some ML experiments in GCP. 

I literally have been trying to spawn ANY instance with ANY GPU for a couple days now. I even wrote a python script to try every possible region because i kept getting 'resource not available' message. 

The only available resources that come back are high/ultra/mega CPU instances. These I don't have any quota for and quota increase requests are automatically declined. So I can't get quota for any of the available ones, and there is no availability for the ones which I have quota.  

Please take my money and give me cloud GPU access. 

Thanks, 

Mike 

[McMaco]

Hello stackoverthrow,

Welcome to Google Cloud Community!

Resource errors occur when you request new resources in a zone that can't accommodate your request due to the current unavailability of Compute Engine resources, such as GPUs or CPUs.

Resource errors only apply to new resource requests in the zone and don't affect existing resources. Resource errors aren't related to your Compute Engine quota. Resource errors are temporary and can change frequently based on fluctuating demand.

Solution

To proceed, try the following:

• Create an instance with a different machine type.
• Create the instance in a different zone.
• Attempt the request again later.
• Reduce the amount of resources that you're requesting. For example, try to create an instance with less GPUs, disks, vCPUs, or memory.

Determine the number of GPUs available in your project by checking the quotas page. If GPUs aren't listed on the quotas page, or you require additional GPU quota, you can request a quota increase. See Request a higher quota limit.

I hope the above information is helpful.