Azure Data Factory Change Data Capture pricing

Question

Hello,
I've searched internet for any reasonable info regarding Change Data Capture feature pricing in Azure Data Factory. The only piece of info I got is this:

But the question is how is this cost calculated? Is it only for this processing time?

CDC is enabled all the time, it works in 30 mins intervals (so basically pulls the data once per 30 minutes) but not sure if the cost is based on the time when CDC is enabled (so 24/7) or for processing time only.

Answer 1

Hi ,

Thanks for reaching out to Microsoft Q&A.

Based on the information highlighted in the first image and ADF's general pricing principles, here’s how the cost calculation likely works for CDC in ADF:

1. Billing is based on vCore usage:

• The CDC process executes on a single node 4 vCore machine.
  • For the public preview, CDC artifacts are billed at General Purpose rates for 4 vCore clusters.

1. Processing Time Matters:
   • The processing time for each CDC execution is logged (in your second image, 1m 24s).
   • Since CDC runs at 30-minute intervals, it's only actively using compute during these runs.
2. Is Billing Based on Execution Time or 24/7?
   • Likely based on execution time only: Azure typically bills for active compute usage rather than idle time. Since CDC is executing in intervals, it should be billed for the sum of the active execution times rather than a full 24 hour period.
   • This follows the model of how Data Flow execution is billed prorated per minute of execution.

How to Estimate Cost?

If your CDC job runs every 30 minutes and each run takes ~1m 24s, in a 24-hour period, it will execute 48 times:

• Total execution time per day = 48 * 1.4 minutes = 67.2 minutes (~1.12 hours).
• Billing is likely at the 4 vCore General Purpose rate per hour, prorated based on this execution time.

To get exact pricing, check the General Purpose pricing for Data Flows (4 vCore) in your Azure region.

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.