Category: Power BI

Phil Seamark has a clever workaround:

Power BI doesn’t yet have a feature that allows end-users to turn on/off the ability to process calculations for visuals on a report page until they are ready. Most of the time, this is perfectly fine – however, in some instances, it can be handy to disable long-running and heavy calculations from running. At the same time filters/slicers are get selected.

The scenario you most likely want to have this control is when your model uses Direct Query mode against large tables in data sources that charge you for query processing. Even if your Direct Query data source does not charge per query, having a user make quick-fire selections over several slicers can potentially saturate a back-end data-source and unnecessarily chew up resources.

Read on to understand how to use calculation groups to do this, as well as the limitations around this solution.

Comments closed

Chris Webb illuminates us:

One of the most common questions I get asked is “How can I use Power BI in DirectQuery mode on top of a REST API?”. This seems like a reasonable thing to do but almost everyone who tries it will fail, and in this post I will explain why.

To answer this question we first of all have to review the two main ways of working with data in Power BI: Import mode and DirectQuery mode. In Import mode data is cached in Power BI’s own internal database and all the DAX queries that are generated by your reports are answered from there. In DirectQuery mode no data is stored inside Power BI; instead, when a report is run and DAX queries are fired off against your dataset, Power BI in turn generates queries against the data source to get the data needed. Most of the data sources that can be used with DirectQuery mode in Power BI are relational databases and so that means Power BI will generate SQL queries to get data from them, but Power BI can also generate queries in other languages too.

Read on for the bad news, although there are some third-party products which can make it work in specific cases.

Comments closed

Imran Burki builds a chart:

Last week we built a Manufacturing Yield dashboard that showed first and final pass yield numbers. In this post, we’re going to introduce the concept of manufacturing defects and build a Pareto chart in Power BI. Unlike Excel, the Pareto chart isn’t out-of-the-box in Power BI. Instead, we must create DAX to build the Pareto. Before we dig into the DAX, let’s talk about why we would create a Pareto in the context of manufacturing and why defects are important to track.

Read on to learn what a Pareto chart is and how you can build the DAX function which gives us the relevant information.

Comments closed

Jon Fletcher shows how to disable query processing in Power Query:

In Power BI Power Query there is an option to enable or disable whether a table is loaded into the report. The option ‘Enable load’ can be found by right clicking on the table. Typically, by default, the load is already enabled.

Click through to see how you can disable query processing or remove a particular query from report refreshes.

Comments closed

Chris Webb shows how you can improve dataflow performance in Power BI after switching to a Premium Per User model:

Over the years I have written a lot about Power BI/Power Query performance but it has always been in the context of loading data direct into datasets, not dataflows. A lot of cool things have been happening in dataflows recently, though, and now that Premium Per User has made Premium features to a much wider audience I thought it would be interesting to look at an example of how PPU can help dataflow performance and specifically how and when the Enhanced Compute Engine can make dataflow refresh faster.

Click through for some interesting findings.

Comments closed