Google pricing: https://cloud.google.com/run/pricing
Terraform resources:

• google_cloud_run_* - some of these might be free

Details

google_cloud_run_service

We will need to extract the following attributes from the Terraform:

• template.spec.containers[*].resources.limits.cpu
• template.spec.containers[*].resources.limits.memory
• template.metadata.annotations["autoscaling.knative.dev/minScale"]
• template.metadata.annotations["run.googleapis.com/cpu-throttling"] (this is true by default)

And add the following usage attributes:

• monthly_requests
• These are used if CPU throttling is enabled:
  • average_request_duration
  • concurrent_requests_per_instance
• These are only used if CPU throttling is disabled.
  • instance_hrs

The cost components will then be calculated as:

• CPU allocation:
  • If CPU throttling is disabled
    • if instance_hrs is specified use that, otherwise use the min scale value * 730 for the instance hours.
    • Multiply the instance hours by the sum of the CPU values to find the CPU usage.
  • If CPU throttling is enabled:
    • The instance hours should be calculated from (monthly requests * average request_duration) / concurrent requests per instance. This should then be multiplied by the sum of the CPU values.
• Memory allocation:
  • Use the same calculation as CPU. This should follow the same logic depending on if CPU throttling is disabled or not.
• Requests:
  • Based on the monthly_requests usage value

The prices for all the cost components are also based on whether the region of the resource is a Tier 1 or Tier 2 region. We might need to define a static mapping for these initially.

google_cloud_run_v2_service

This uses the same logic as the google_cloud_run_service resource but the attributes are slightly different:

• template.containers[*].resources.limits.cpu
• template.containers[*].resources.limits.memory
• template.scaling.min_instance_count
• template.containers[*].resources.limits.cpu_idle

google_cloud_run_v2_job

We will need to extract the following attributes from the Terraform:

• template.template.containers[*].resources.limits.cpu
• template.template.containers[*].resources.limits.memory
• template.task_count

Pricing again depends on if the region is a tier 1 or tier 2 region.

And add the following usage attributes:

• monthly_job_executions
• average_task_execution_mins

The cost components will then be calculated as:

• CPU allocation: monthly job executions * task count * task_execution_mins * cpu
• Memory allocation: monthly job executions * task count * task_execution_mins * memory

These resources can be added as free resources:

• google_cloud_run_service_iam_policy
• google_cloud_run_service_iam_binding
• google_cloud_run_service_iam_member
• google_cloud_run_domain_mapping
• google_cloud_run_v2_job_iam
• google_cloud_run_v2_service_iam