Blog Post

Azure Data Factory Blog
1 MIN READ

Data Flows in Azure Data Factory Now Support Reserved Instance Pricing

Mark Kromer's avatar
Mark Kromer
Icon for Microsoft rankMicrosoft
Feb 09, 2021

You can now purchase 1-year or 3-year reservations of Data Flows from the Azure Portal and receive up to 30% off the pay-as-you-go option for General Purpose and Memory Optimized compute options in Azure Data Factory.

 

Read more about the way to estimate your hourly usage and how to enter your reservations at these documentation links:

https://docs.microsoft.com/en-us/azure/data-factory/data-flow-reserved-capacity-overview

https://docs.microsoft.com/en-us/azure/data-factory/data-flow-understand-reservation-charges

 

Here's an example: If you typically use 32 cores of Memory Optimized data flow compute per hour, you can add a reservation for those 32 cores and receive a discount from the pay-as-you-go pricing based on the number of years that you set for your reservation. If you use 64 cores of Memory Optimized Azure IRs in data flows for an hour, then you will receive the discount on the first 32 cores, then pay the normal rate for the remaining 32 cores. If you use 16 cores in the next hour, then you pay for the entire 32 cores of your reservation.

Updated Mar 18, 2021
Version 2.0
  • nckanth's avatar
    nckanth
    Copper Contributor

    By Reserving Instance do I reduce cluster startup time ?

  • Pushkar1815's avatar
    Pushkar1815
    Copper Contributor

    By using Pay As You Go in azure, Will I be able to connect to PostgreSQL DB in mapping data flow which is implemented inside azure data factory.

     

    I want to connect to an On-premise postgreSQL DB at source and then transform the data in mapping data flow, once done then again send the transform data back into On-premise PostgreSQL DB. If you see here, the options for PostgreSQL are not active. So does that mean that I need Upgrade to PAY AS YOU GO Subscription.