Creating a 80GB A100 instance on Google Cloud

1. Sign up for Google Cloud.

2. Set up a billing account.

3. Create a new project called “alpaca” and copy and paste the generated project ID.

4. Go to https://console.cloud.google.com/iam-admin/quotas and request a quota increase for compute.googleapis.com/nvidia_a100_80gb_gpus in us-east4. Note that this may take several hours, sometimes even a couple of days.

5. Install the gcloud command line tool.

6. Run gcloud auth login to login using your browser.

7. Run gcloud init . Select the project you created and the zone us-east4-c.

8. Run this to create an instance:

   gcloud compute instances create \

   alpaca \

   --project=alpaca \

   --zone=us-east4-c \

   --machine-type=n1-standard-8 \

   --accelerator type=nvidia-tesla-t4,count=1 \

   --boot-disk-size=1024GB \

   --image-project=deeplearning-platform-release \

   --image-family=common-cu113 \

   --maintenance-policy TERMINATE \

   --restart-on-failure \

   --scopes=default,storage-rw \

   --metadata="install-nvidia-driver=True"

9. Wait a few minutes, and then log in to the new instance:

   gcloud compute --project=alpaca ssh --zone=us-east4-c alpaca

10. Remember to switch it off when you’ve finished:

    gcloud compute instances stop alpaca