Use field parameters to allow users to switch between distinct counts and approximate distinct counts

Connections are now easier to manage

Get the date and time that measures, columns and other objects were last changed

Storing images and other files in OneLake avoids the need for complex workarounds

UI based tools for Power BI developers in Fabric notebooks

Drill to detail for a measure in a Power BI report

Detail Rows Expressions can sometimes offer an alternative to user defined aggregations

Excel Copilot can now get data direct from Power BI semantic models

DAX UDFs can be used my M365 Copilot for complex data analysis

A simple example of using an Operations Agent to watch for broken visuals in a Power BI report

Using DirectQuery on Fabric Warehouse or the SQL Endpoint of a Lakehouse is great for large data volumes

The value of this new book lies it its worked examples of how to solve complex DAX problems

Disabling V-Order on staged data makes staging faster

More parallelism does not always result in faster performance and can cause errors

How to control concurrency within a dataflow when partitioned compute is used

A simple example of how increasing concurrency can improve dataflow refresh performance

Refreshing too many semantic models at the same time on a capacity can cause this error

Configure your semantic model so it connects to a Shareable Cloud Connection immediately after publishing

Steve Jones recently posted an update about SQL Saturday’s status, and it includes some news we need to talk about: However, this year the number may stagnate or even decline slightly. Running events…

The FabricAI.Prompt can be used to generate CSV data

Power BI now shows warnings about broken measures and other items in its refresh history

A workaround to allow you to create multiple tables in a Direct Lake semantic model linked to the same table in OneLake

Dataflows Gen2 can now generate Excel files with tables and charts

Partitioned Compute lets you parallelise tasks inside a Dataflow

I have been spending a lot of time recently investigating the new performance-related features that have rolled out in Fabric Dataflows over the last few months, so expect a lot of blog posts on th…

How strict and eager evaluation can affect parallelism in DirectQuery mode

Filtering on 0 can be bad for performance if you're not careful

Large images can make reports slow

Geocoding could be making your report slow

Large resultsets can be slow to read

See when a DirectQuery report is slow because there are limited connections back to the source

How Performance Analyzer data exports can be visualised in Power BI using a custom visual

How report filters and slicers influence how Copilot answers questions

AI Instructions can force Copilot to get answers from the semantic model instead of the report

Significant performance improvements are possible but not in all cases

Build permissions are not security!