Discrepancy in Billable Hours Calculation Between Workspaces Cost Optimizer, CUR Data and cloudwatch metrics

[ramesh106]

In our deployment of the Workspaces Cost Optimizer across few aws accounts, we conducted a dry run on the workspaces. During this process, we observed a discrepancy in the billable hours for a workspace set to autostop mode.

In the aggregated dry run CSV report, a workspace displayed 526 billable hours. Considering the dry run parameter true, which should cover the billable hours from April 1 to April 20, the maximum possible billable hours would be 480 (20 days * 24 hours/day).

Upon inspecting the CUR (Cost and Usage Report) data for the same workspace, it indicated 444 billable hours, the accurate figure for billing purposes.

This significant disparity raises concerns about the calculation methods employed by the Workspaces Cost Optimizer. We need to ascertain whether this is due to a bug in the code or a discrepancy between the calculation mechanisms of the optimizer and the CUR data.

Additionally, it's worth noting that the hours are not matching exactly between CloudWatch metrics viewed through console and the Workspaces Cost Optimizer calculated billable hours, further highlighting the need for resolution.

[G-Lenz]

Thank you for bringing this to our attention, we have added this to our backlog and will be looking into this discrepancy.

[G-Lenz]

Our solution uses the UserConnected CloudWatch metric to determine the Workspace usage. Are you able to confirm what the usage was for that time period for the affected workspaces in CloudWatch and see if it matches up with what the solution is reporting?

[G-Lenz]

Closing due to inactivity, please respond and re-open if this issue persists.