Details
-
Story
-
Resolution: Done
-
Normal
-
None
-
None
-
False
-
None
-
False
-
COST-2820 - Walk: cost of running OpenShift
Description
Summary
Calculate the cost of unutilized node capacity during the OCP on Cloud SQL calculation, using node cost from cloud source
- OCP on Cloud flow will be slightly different in that we’re grabbing the cost directly, instead of just calculating the usage difference, that is already covered in the “on prem” flow
- Alongside OpenShift on AWS/Azure/GCP summary SQL, run new Trino SQL to determine unutilized capacity
can probably be generalized single SQL file that operates against a table depending on underlying cloud source
Per node, per day
- Insert new record in table
- Sum the cost for everything running on that node
- Subtract the summed cost from the total node cost reported by the cloud source to determine the cost of unused capacity
- Set the project name based on the node’s role determined above (platform idle capacity) for master/infra and idle capacity for worker nodes
- This will give us a line item for each node per day that has a simulated project name for idle capacity cost