Reload only one year data using dataflow while remianing past years data is same

[anusha_2023]

I am loading General Ledger table into Fabric Lakehouse using dataflow and everyday and published Power BI model is refreshing only 2024 data through incremental refresh while 2020, 2021, 2022, 2023 remain same. Now I want to make change in the dalow too to reload data for only 2024 while other years remiain constant or no need to reload. How to achive this scenario using dataflow. I know that it is easy with notebook but I am using Onpremises gateway so I need to load data using dataflow. Please suggest the ideas.

[lbendlin · Accepted solution]

You may have wanted to mention that you need this for DF Gen 2.  The concept there is slightly different. Note the limitations.

Incremental refresh in Dataflow Gen2 - Microsoft Fabric | Microsoft Learn

[anusha_2023]

I have a dataflow in the Fabric Workspace which is loading 2021 to 2024 all the time. I want to add changes to the dataflow where 2021 to 2023 need to be constant and 2024 need to be updated for each data load.

[lbendlin]

Configure your incremental refresh to keep the last five years and refresh the last two years?

[anusha_2023]

How can I configure the Incrementa refresh inside dataflow to keep first 3 years constant and 2 years reload for each run. Could you please share me the steps to work on.

[lbendlin]

Note that you get one more year partition "for free"  because the 5 years is counted back from Today() and therefore includes a small part of 2019. That partition will drop off on Jan 1st.

[anusha_2023]

I could not see the option for Incremental Refresh in Fabric dataflow Gen2. 

[lbendlin]

You said you are already using incremental refresh?  Please provide a more detailed explanation of what you are aiming to achieve. What have you tried and where are you stuck?