Register now to learn Fabric in free live sessions led by the best Microsoft experts. From Apr 16 to May 9, in English and Spanish.
This is going to sound dumb, but where are imported datasets technically stored in the Power BI Service? Please do not answer with "see this link" unless accompanying a complete answer.
Answers that don't count:
In Azure
In the PBI Service
In the PBI Cloud
With dataflows the answer is very easy to find - the Azure datalake Gen2. And according to https://docs.microsoft.com/en-us/power-bi/admin/service-admin-power-bi-security imported datasets are sent to Azure Blob Storage. But if that were true, I don't understand these things
#1 - If datasets are in blobs, I should be able to access them like any other blob storage in Azure, right?
#2 - Alds gen2 is built on azure blob storage, and if I can access dataflows directly from adls gen2 datalake, couldn't the blob underneath be exposed?
#3 - If our Premium capacity has a 100TB limit then it would have to take the total of what was in Azure Blob storage plus the Azure data lake Gen2, but I can't find any such breakout in admin settings or API's.
This should be so simple, but I'm struggling to understand what a dataset really is.
Solved! Go to Solution.
Hey @PBI4LIFE ,
sorry I thought you mean when actively used.
Yes, the security whitepaper mentions that for import and composite data models the data is saved either in Azure Blob Storage or in Azure SQL Databases. But they don't give further details. When the data is actively used "Power BI loads actively processed data into the memory space of one or more service workloads":
https://docs.microsoft.com/power-bi/guidance/whitepaper-powerbi-security#data-handling
For me all of that sounds like they use a technique similar to Azure Synapse Analytics.
Hey @PBI4LIFE ,
a Power BI dataset is based on SSAS tabluar technology or as it's in the cloud Azure Analysis Services.
If you want to get an official statement you can quote Chris Wade:
"Azure Analysis Services (AAS) and SQL Server Analysis Services (SSAS) are based on mature BI engine technology used by countless enterprises. The same technology is also at the heart of Power BI datasets"
https://powerbi.microsoft.com/en-us/blog/power-bi-as-a-superset-of-azure-analysis-services/
That also explains why you can connect to premium datasets with XMLA endpoint like to a usual SSAS.
If I answered your question I would be happy if you could mark my post as a solution ✔️ and give it a thumbs up 👍
Best regards
Denis
Blog: WhatTheFact.bi
Hey @selimovd ,
Thanks for getting back to me. When I look at this documentation from Microsoft it says
"data that is uploaded from users is typically sent to Azure Blob Storage"
https://docs.microsoft.com/en-us/power-bi/admin/service-admin-power-bi-security
I thought only the data in-memory was being loaded to SSAS/Tabular/Vertipaq/idk.
So our P1 node has 8 cores and 25 GB RAM and 100 TB storage limit. My understanding is data is loaded to the "SSAS" Tabular in-memory model represented by that 25GB of RAM. And when not being used, it's evicted. I guess my question is - where is it being evicted to/ pulled from when needed again?
Anthony
Hey @PBI4LIFE ,
sorry I thought you mean when actively used.
Yes, the security whitepaper mentions that for import and composite data models the data is saved either in Azure Blob Storage or in Azure SQL Databases. But they don't give further details. When the data is actively used "Power BI loads actively processed data into the memory space of one or more service workloads":
https://docs.microsoft.com/power-bi/guidance/whitepaper-powerbi-security#data-handling
For me all of that sounds like they use a technique similar to Azure Synapse Analytics.
Covering the world! 9:00-10:30 AM Sydney, 4:00-5:30 PM CET (Paris/Berlin), 7:00-8:30 PM Mexico City
Check out the April 2024 Power BI update to learn about new features.