Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!
Dear eveyone,
When I read Dataset modes, there are some sentences make me so much confused:
https://learn.microsoft.com/en-us/power-bi/connect-data/service-dataset-modes-understand#import-mode
I have watched this video to figure out what is the actual flow of the process here:
https://www.youtube.com/watch?v=kpAZD7AeRQw
Also read this blog:
https://blog.crossjoin.co.uk/2022/11/21/how-much-data-can-you-load-into-power-bi/
Until now I imaged that: Datasource (Like Excel) in SSD ---> Stored Disk (The Power BI file which has it owned size - I don't know this size is equal to the Datasource before?) ----> Memory (RAM) (What is the situtation the data loaded into RAM? And when it published to PBI Service where is the RAM to use?)
Another question is what is the stage 10-times compression? and 20% reduction size from what stage?
This really make me confused so much, please help me the answer.
Many thank to you guys ❤️
HI @TungNguyen_19,
Normally power bi will getting data from external data source and convert them to power bi data model tables. (power bi use Analysis Service tabular model to store these table and records and they may be compressed or expanded during these steps)
SQL Server Analysis Services Tabular Fundamentals (mssqltips.com)
If you want to clearly check how these table and records stored, you can also try to use SSMS or excel directly to connect to power bi data model instance and use T-Sql query to view them.
Connect to Power BI Desktop Model from Excel and SSMS (biinsight.com)
Regards,
Xiaoxin Sheng
Hi @v-shex-msft
I still don't understand what are differences between "When loaded from disk to memory - 10 times compression" vs "Storage size on disk can achieve a 20% reduction from the compressed size." - Which one is from Hard drive (like SSD or Data sources) to Tabular model and which one is from Tabular model to In-memory?
Is it true that there are 2 stages here as I mention above?
User | Count |
---|---|
140 | |
113 | |
104 | |
77 | |
65 |
User | Count |
---|---|
135 | |
116 | |
101 | |
71 | |
61 |