New Machine types a2-ultragpu (1g, 2g, 4g, 8g)

[github-actions[bot]]

Machine types 'machinetypes.csv' changed

Added: a2-ultragpu-1g;12;False;170.00;; a2-ultragpu-2g;24;False;340.00;; a2-ultragpu-4g;48;False;680.00;; a2-ultragpu-8g;96;False;1360.00;;

Deleted:

Todo:

• [x] Check changes
• [x] Edit title of this issue
• [ ] Run build/skus.pl workflow to export SKUs from the Google Cloud Billing API
• [ ] Check if machine type is present in 'build/skus.csv'
• [ ] Add or remove mapping in build/mapping.csv
• [x] Add or remove the machine type in build/gcp.yml
• [ ] Add or remove the machine type with mapping in build/pricing.pl
• [ ] Check if machine type gets SUD and edit build/pricing.pl
• [ ] Create a test for machine type in t/test.sh
• [ ] Test cost calculation with machine type
• [ ] Build new pricing information file 'pricing.yml'
• [ ] Run Open Graph action in Cyclenerd/google-cloud-compute-machine-types
• [ ] Run Build action in Cyclenerd/google-cloud-compute-machine-types

[Cyclenerd]

Not official released: https://cloud.google.com/compute/docs/accelerator-optimized-machines#a2_machines

[Cyclenerd]

Docu: https://cloud.google.com/compute/docs/accelerator-optimized-machines#a2-ultra-vms

[Cyclenerd]

Can not find SKU for NVIDIA A100 80GB (nvidia-a100-80gb). Currently I calculate the price with the normal A100 with 40GB. Price is therefore too low!

I can't start an instance either. Currently there is this quota:

Operation type [insert] failed with message "Quota 'NVIDIA_A100_80GB_GPUS' exceeded. Limit: 0.0 in region us-central1."

Look at it again later.

Instances are in gcp.yml: https://github.com/Cyclenerd/google-cloud-pricing-cost-calculator/commit/4d6de1dfac49df019b1643974fc548e5445ecdb6

[Cyclenerd]

SKU: Nvidia Tesla A100 80GB GPU running in

[Cyclenerd]

Done: https://github.com/Cyclenerd/google-cloud-pricing-cost-calculator/commit/ab2ac6c91f48e17061f0ed6d1fa9f13dd454eb84