Lineage Limitation on Wide Power BI Semantic Models & Built-in Classification Rule Sets

Hi Nai,

Regarding the column count limit, there isn't a specific limit set, but we have noticed issues when processing large amounts of data at once, especially when the count exceeds 200 columns. It is recommended to break down the data and try processing 150-200 columns at a time though 200 should work fine. You can create sub-datasets or rules to manage this.

 Built-in Classification Rules Currently, it is not possible to permanently delete the built-in classification rules. However, you should create custom classifications according to your requirements.

 Scan Strategies To avoid full scans and unnecessary filters, you can use scoped and incremental scans. Additionally, you can use exclusion methods to ensure that certain assets or classifications are not applied during a scan. 

Automation details : https://learn.microsoft.com/en-us/purview/data-gov-best-practices-automation#streaming-apache-atlas 

Feel free to tell if it doesn't help or you need more information.