We have come up with t-shirt sizes for dedicated instances where each size maps to specific instance types for compute & database, so that we can control our costs, charge customers accordingly, and standardize across the fleet. That configuration is driven from TF. It's already effected for AWS in production so all instances are on a t-shirt size relevant to each customer. But we are yet to make that change for GCP. For GCP, I think the changes are in place, but will need to be validated, tested and integrated end to end. See the screenshot for the per t-shirt size configuration:
We have come up with t-shirt sizes for dedicated instances where each size maps to specific instance types for compute & database, so that we can control our costs, charge customers accordingly, and standardize across the fleet. That configuration is driven from TF. It's already effected for AWS in production so all instances are on a t-shirt size relevant to each customer. But we are yet to make that change for GCP. For GCP, I think the changes are in place, but will need to be validated, tested and integrated end to end. See the screenshot for the per t-shirt size configuration: