Hi ,
Thanks for reaching out to Microsoft Q&A.
This unexpected Azure Databricks VM pricing behavior could be due to several reasons. Here are some common reasons & possible explanations:
- Spot Pricing or Spot Eviction Discounts
- Azure may be applying Spot VM pricing dynamically if Databricks clusters are using Spot VMs.
- Even if the customer didn’t explicitly opt for Spot VMs, Databricks can leverage them in autoscaling scenarios.
- Check if the VM is marked as a Spot instance in the Azure portal.
- Azure Hybrid Benefit (AHB)
- If the customer has Software Assurance (SA) or eligible Windows Server/SQL Server licenses, Azure Hybrid Benefit could be applied automatically.
- However, this typically applies more to Windows-based VMs rather than Linux.
- Promotional or Discounted Pricing
- Microsoft occasionally applies regional promotions or commitment-based pricing even if RI is not explicitly purchased.
- Contact Azure Support to verify if a temporary discount or an Azure Consumption Commitment (MACC) incentive is being applied.
- Enterprise Agreement (EA) or Negotiated Pricing
- If the customer is on an Enterprise Agreement (EA) or CSP subscription, negotiated pricing or bulk discounts could be in effect.
- The actual cost may differ from the public Azure Pricing Calculator.
- Databricks-Specific Optimizations
- Azure Databricks manages VM resources internally and might be applying reserved capacity behind the scenes.
- Some pricing adjustments happen at the Databricks level, not always at the subscription level.
- Potential Cost Anomaly or Billing Error
- If none of the above apply, it could be a billing discrepancy.
- Raise a support ticket with Azure to validate the applied pricing logic.
Next Steps:
- Check if the VM is Spot or has any hidden discounts.
- Look into Hybrid Benefits or EA Pricing.
- Verify with Microsoft Support if there's an automatic discount.
- Monitor the next billing cycle to ensure it's not a temporary adjustment.
Please feel free to click the 'Upvote' (Thumbs-up) button and 'Accept as Answer'. This helps the community by allowing others with similar queries to easily find the solution.