Hi,
We have a small project on GCP with Compute-optimized C2D machines (8 vCPU + 32 GB RAM) that was budgeted to cost ≤$180/month based on our understanding of CUDs. However, despite having active commitments, our monthly costs consistently exceed $350 for Compute Engine resources. After a year of frustrating support tickets, we need expert community assistance.
Based on our CUD SKU prices:
- C2D Cores: $0.013303 per vCPU hour
- C2D Memory: $0.001781 per GB hour
For our configuration (8 vCPUs and 32 GB RAM):
- 8 vCPUs × $0.013303 = $0.106424 per hour
- 32 GB RAM × $0.001781 = $0.057392 per hour
- Total hourly cost = $0.163816
- Monthly cost (730 hours): $119.59
we assumed that our calculated commitment cost ($119.59) should be drastically lower than our actual monthly bill ($350+).
Questions we need answers to (or help with) :
-
Is the Compute-optimized C2D commitment supposed to be a standalone discount, or does it require purchasing an additional “regular” CUD to gain the benefit?
-
Would purchasing a different 8 vCPU + 32 GB RAM CUD in us-central1 reduce our cost, or would these duplicate commitments we already have?
-
Why is there such a significant gap between our calculated committed price (~$120/month) and actual billing ($350+/month)?
We’ve opened multiple tickets with GCP Billing Support over the past year, but each agent has provided contradictory information. The support team doesn’t seem to understand how their own CUD products work for Compute-optimized machines.
Screenshot
Additional Context
- Region: us-central1
- Machine type: Compute-optimized C2D
- Configuration: 8 vCPUs, 32 GB RAM
- Active Commitments: Compute-optimized C2D Cores & Memory (3-year term)
We’re seeking expert community advice as we’ve exhausted official support channels. Any insights on why our actual costs are nearly triple our calculated commitment costs would be greatly appreciated.
Thank you for the expected help!