Azure Databricks · FinOps Profile

Azure Databricks Finops

FOCUS-aligned FinOps for Azure Databricks: dual-meter pricing where DBUs (Databricks Units) are billed by Databricks via Azure Marketplace and the underlying Azure VM, storage, and network are billed by Azure. Reserved DBUs (DBCU) and Azure Reservations stack for committed-use savings.

Azure Databricks Finops is the FinOps profile for Azure Databricks on the APIs.io network, aligned with the FinOps Foundation Framework.

It defines 6 billable meters, billed in USD, on a monthly cycle, and pricing category pay-as-you-go + committed use.

The profile maps 8 FOCUS columns for cost-allocation reporting.

Tagged areas include Analytics, Lakehouse, Spark, FinOps, and FOCUS.

Category: Data Analytics Pricing: Pay-As-You-Go + Committed Use Billing: Monthly FOCUS v1.3
AnalyticsLakehouseSparkFinOpsFOCUS

Framework Alignment

Framework
FinOps Foundation Framework
Data Spec
FOCUS v1.3

Charge Categories

UsagePurchaseTaxAdjustment

FOCUS Columns

BillingCurrency
USD
ChargeCategory
Usage
InvoiceIssuerName
Microsoft Corporation
PricingCategory
Pay-As-You-Go
ProviderName
Microsoft
PublisherName
Microsoft Corporation
ServiceCategory
Data Analytics
ServiceName
Azure Databricks

Meters

dbu_consumption
Unit: DBU
Databricks Unit consumption (per second) by workload type
vm_compute_hours
Unit: instance-hour
Underlying Azure VM time (billed separately by Azure)
serverless_compute
Unit: DBU
Serverless SQL / Model Serving / Jobs compute charged as DBUs without separate VM line
storage_capacity
Unit: GB-month
Underlying ADLS Gen2 / managed storage
model_serving_queries
Unit: query
Inference queries against Model Serving endpoints
vector_search_capacity
Unit: endpoint-hour
Vector Search endpoint storage and throughput units

Sources