how dataset is being compressed in Power Bi Service?

[pennyhoho117]

When a pbix is using import mode, and there is a limitation: 

Semantic model size limit 1GB,

First, May i know what is the defination of 'Semantic model'

For example, when we publish a report to Power BI Service,

two different object: Report and Semantic Model would be generated.

then when we click into the semantic model, there may be 10 tables in a Semantic Model, 

So the limitation 1GB is count for all 10 tables in the semantic model, or

the limitation 1GB is count each one table in the semantic model?

Second,  i know the compression is using VertiPag Engine in Power BI cloud, 

so after publish the report to the power BI service, each time when refreshing data, would compress the VertiPag in where and when and how?

[v-sgandrathi · Accepted solution]

Hi @pennyhoho117,

Defination of semantic model:

The Semantic Model in Power BI is the underlying data model that stores imported data, relationships, calculated columns, measures, and hierarchies

The 1GB semantic model size limit applies to the entire model, meaning all tables combined. If your semantic model contains 10 tables, their combined size after compression must fit within the 1GB limit.

As per my knowledge after publish the report to the power BI service, each time when refreshing data, would compress the VertiPag inside the powerBI Service when the data is initially loaded or refreshed. 
When:

 When you publish a report using Import Mode, the data is compressed using VertiPaq before being stored in Power BI Service.

Each time the dataset is refreshed, Power BI reloads and re-compresses the data using the VertiPaq engine to optimize performance.

How:

VertiPaq compresses data column-wise instead of row-wise, meaning that it stores unique values efficiently and applies advanced encoding techniques.

If this post was helpful, please consider marking Accept as solution to assist other members in finding it more easily.

If you continue to face issues, feel free to reach out to us for further assistance!

Thankyou for reaching out to the Microsoft Fabric Community Forum.