Forum Discussion
New Power BI Embedded pricing model - page render
I am not sure I understand totally the term *page render* in new Power BI Embedded pricing model.
We need to measure how often our clients load Power BI in order to charge them properly. But we still got the confusion how page render is counted under the hood.
From the definition:
A page render is counted any time Power BI visuals are loaded on a page. A page refresh counts as a page render, as does any other page interactivity, like slice and dice, filtering, etc. Typically, multiple queries running on the backend cores translate to a page render.
How we can understand Page Render in both cases Import and Direct Query mode for Report Embedding.
Scenario 1 with import mode:
Assume one page in our application *embeds an report with import mode*. This report has 2 tabs inside and each tab has 5 visual. So, this report has 10 visuals in total.
How many page render if user visits this page: 1 render (based on the whole report), or 2 (based on 2 tabs), or 10 (based on 10 visuals).
If user switches between two tabs in the report, is there any page render is counted?
Now assume there is datetime slicer in the tab 1, if user use this slicer to do filtering. One render is counted on this case?
Scenario 2 with direct query:
The same with the scenario 1 with direct query report.
8 Replies
Many years after the original question was posted, I am still looking for an answer as the documentation is not clear. Would any one from Microsoft team respond and clarify what counts as page render? In my case, I have an embedded report with A1 capacity that is connected to Direct Query mode which refreshes every 10 second. I wonder if any page refresh is counted only as one render or the count of renders = the number of visuals/filters * 10 per each page refresh?
Thanks for your reporting, we'll consult the product team for more clarfication about the page renders.
By the way, as to peak hour page renders, when exceeding the maxium, it doesn't mean your reports won't be rendered any more, see below clarification.
- There're no hard limit. In peak time when there are too many requests exceeding the limit, request will get queued and end customer will experience performance degradation (page render will take more time)
Thanks Eric_Zhang
I understand your point, it would be great if we have clear understanding on page render on the new Power Bi Embedded.
The response from the product team is
“page renders” are just a rough guideline, not an actual resource or limit. Therefore, it is not productive to get too detailed about what counts as a page render and what doesn’t.
The only limited resources in capacity are CPU, memory, and DirectQuery queries. Slicing, filtering, etc. will take up CPU and memory. If the dataset is DirectQuery, it will also take up DirectQuery queries.
