Data Factory
Azure Data Factory — commitment brief
Data sourced: March 2026. Verify current figures at the Azure Pricing Calculator.
Scope note
ADF has a split commitment posture. Reservations apply only to Data Flow compute (Mapping Data Flows running on Azure Integration Runtime). Pipeline orchestration activity runs, Copy activity DIU-hours, self-hosted IR execution, and all other pipeline activities have no commitment instrument available and are always billed pay-as-you-go.
Coverage summary
Azure Data Factory supports reservations specifically for Data Flow compute (vCore-hours) on General Purpose and Memory Optimized compute types. These are purchased per compute type, core count, and Azure region — not per factory or integration runtime. All factories in scope automatically receive the benefit. Change Data Capture (CDC) resources are covered by the same Data Flow reservation. The Compute Savings Plan may partially apply to Data Flow compute where it runs on eligible Azure IR infrastructure, but the dedicated Data Flow reservation is the stronger and more targeted instrument.
What is covered: Data Flow compute (vCore-hours) on Azure Integration Runtime — General Purpose and Memory Optimized tiers. CDC resources at General Purpose rates.
What is not covered: Pipeline orchestration activity runs, Copy activity DIU-hours, self-hosted IR execution, storage (managed disk and blob) used during Data Flow execution and debugging, networking charges.
Azure Hybrid Benefit
Not applicable. Azure Data Factory is a PaaS service with no SQL Server or Windows Server licensing component.
SKU / tier coverage
General Purpose (vCore)
✅ Yes
⚠️ Partial
~25%
~35%
$0.274 PAYG → $0.205 (1yr) → $0.178 (3yr) per vCore-hr. Also covers CDC resources.
Memory Optimized (vCore)
✅ Yes
⚠️ Partial
~25%
~35%
Required for complex joins/aggregations. Purchased separately per core count.
Pipeline orchestration runs
❌ No
❌ No
—
—
Billed per activity run at PAYG. No commitment instrument available.
Copy activity (DIU-hours)
❌ No
❌ No
—
—
DIU consumption always PAYG. Optimise by right-sizing DIU count, not committing.
Self-hosted IR
❌ No
❌ No
—
—
Customer-owned compute. Optimise at the VM layer via VM reservations instead.
Reservation vs. Compute Savings Plan
Data Flow Reservation (recommended): Purpose-built for ADF. Scoped to compute type (General Purpose or Memory Optimized) + core count + region. Delivers up to ~35% at 3yr. Best for teams with consistent, predictable Data Flow execution. Available as 1-yr and 3-yr terms.
Compute Savings Plan: Applies to ADF Data Flow compute only where it runs on eligible Azure IR compute infrastructure. Offers flexibility across regions and compute types but a lower discount ceiling than the dedicated Data Flow reservation. Better suited as a secondary instrument or for variable workloads.
Commitment risk flag
ADF Data Flow reservations are use-it-or-lose-it per hour. If your Data Flows don't execute the full hour matching the reserved core count, the unused portion is wasted for that hour — it does not roll forward. Reservations apply only when matching Azure IR resources are actively running. Idle or intermittent pipelines will underutilise the commitment. Size reservations to your consistent baseline execution, not your peak.
Regional availability
ADF Data Flow reservations are purchased per region and are available in most commercial Azure regions. General Purpose vCore (product ID DZH318Z0C1BN) is broadly available across commercial regions. Memory Optimized vCore (product ID DZH318Z0C12B) has more limited regional availability — it is not available in East US or East US 2, but is available in West US 3, France South, Germany West Central, and several other regions.
Notable exclusions: Memory Optimized reservations are not purchasable in East US or East US 2. Azure Government and Azure China regions are not supported for ADF Data Flow reservations.
⚠️ Availability varies by SKU and region. Always verify in the Azure portal or via the Azure Retail Pricing API before purchasing.
Archera
Azure Data Factory Data Flow reservations are within Archera's commitment management scope. Given the hourly use-it-or-lose-it risk on ADF reservations — particularly for bursty or intermittent pipeline schedules — Archera's GRI/GSP wrapper and utilisation monitoring are especially valuable here to protect against commitment waste.
Sources
⚠️ vCore-hour rates ($0.274 GP PAYG, $0.205 1-yr, $0.178 3-yr) are East US region indicative figures as of March 2026. Rates vary by region. Always verify with the Azure Pricing Calculator.
Last updated
Was this helpful?