When initially configuring incremental refresh in Power BI Desktop, the parameters are used to filter only a small period of data to be loaded into the model. Table data is filtered by using Power Query date/time parameters with the reserved, case-sensitive names RangeStart and RangeEnd. With incremental refresh, the service dynamically partitions and separates data that needs to be refreshed frequently from data that can be refreshed less frequently. If the table is large, say with tens of millions of rows or even more, a refresh for that table can take a long time and consume an excessive amount of resources. That single partition contains all rows for that table. When you publish a Power BI Desktop model to the service, each table in the new dataset has a single partition. When published, the service automatically applies those policies with each refresh. Easy setup - Incremental refresh policies are defined in Power BI Desktop with just a few tasks.Enables large datasets - Datasets with potentially billions of rows can grow without the need to fully refresh the entire dataset with each refresh operation.Resource consumption is reduced - Less data to refresh reduces overall consumption of memory and other resources in both Power BI and data source systems.Queries to source data run faster, reducing potential for network problems to interfere. Refreshes are more reliable - Long-running connections to volatile data sources aren't necessary.

Refreshes are faster - Only the most recent data that has changed needs to be refreshed.Fewer refresh cycles for fast-changing data – DirectQuery mode gets the latest data updates as queries are processed without requiring a high refresh cadence.With incremental refresh and real-time data: An incremental refresh policy to partition the table, refreshing only the most recent import partition(s), and optionally leveraging an additional DirectQuery partition for real-time data can significantly reduce the amount of data that has to be refreshed while at the same time ensuring that even the latest changes at the data source are included in the query results. For most datasets, this is one or more tables that contain transaction data that changes often and can grow exponentially, like a fact table in a relational or star database schema. Incremental refresh extends scheduled refresh operations by providing automated partition creation and management for dataset tables that frequently load new and updated data.

Refresh option on db solo