Exam PL-300 topic 2 question 16 discussion

I agree with the answer. B. No

I think query folding can push the filter into the query so A yes

❌ Why the Answer Is No: Even though filtering in Power Query can reduce data size, this approach is inefficient in this context unless the filtering is done before importing the data. Here's the critical issue: If you first import the full table into Power BI and only then filter it, the entire 100+ million rows are loaded before the filter is applied. This defeats the purpose of sampling — it strains resources and takes too long. The answer is No because the filter is applied after import, which doesn’t meet the goal of sampling during development.

Answer: A. Yes ✅ Explanation: The goal is to import a sample of the data from a very large SQL Server table (100+ million records) during development — not the full dataset. Why this solution works: Power Query Editor allows you to add a filter step to restrict the amount of data loaded (e.g., filtering on date, order ID, or top N rows). While this filtering happens after import, Power BI is smart enough to fold the filter back to SQL Server if query folding is supported, meaning only the filtered rows are pulled from the database. Query folding: If your filter step can be translated to SQL (like a WHERE clause), Power BI pushes it to SQL Server, making the data load efficient even with massive tables. Best Practice Tip: For better control, you could also: Use a SQL statement in the advanced options during import to fetch a subset directly (e.g., SELECT TOP 1000 * FROM Orders).

Giải pháp được đề xuất ("nhập bảng rồi mới lọc") không đạt được hiệu quả lấy mẫu nhanh chóng vì nó vẫn cố gắng tải toàn bộ dữ liệu ban đầu.

"During the development process, you need to import a sample of the data from the Order table.". This approach meets this goal thanks to query folding, as after the filtering step, the native query sent to the data source will be modified to import only the rows that meet the condition.

Filtering ensures that we select only a manageable number of rows such as 100, 000 that will help test what we want to analyze rather than having 100 million rows

B. It stated "From Power Query Editor, you import the table and then add a filter step to the query." Can't do this. If imported through and SQL and not the table, then it is probably a Yes.

Power Query Editor allows filtering data at the query level, which means only the filtered data will be imported into Power BI, reducing the data load

In Power Query Editor, when you add a filter step to the query, only the filtered subset of data is loaded into the Power BI model. This approach allows you to work with a smaller sample of the data during development while maintaining the ability to later remove or adjust the filter for the full dataset.

A is correct same results chat gpt returned me

you can select the required data in the sql query itself instead of importing all data and then apply filter in power query

B is correct as to minimize the data size while in the development phase

B is the answer you can only apply the filter at the source through SQL for example no after loading the entire table.

During the development process, you need to import a sample of the data from the Order table.: if you choose A you import the whole table and then filter it, so you should adjust the sql statement

Correct answer is Yes , this solution works because adding a filter step in Power Query Editor limits the data imported, allowing you to work with a sample of the table during development.

When the data is imported to power query, then filtered, query folding is possible, hence the filtering is done by the source before retrieving it back to power query. I tested this 1) import a table from SQL to power query. 2) saved the PBIX file. 3) Checked the size of the PBIX file. 4) Open power query and add a filter step to the table. 5) Save the PBIX file. 6) Checked the size of the PBIX file and noticed that the file size actually decreased. This shows that only the required data is imported If the entire data from the SQL is imported to power query, then only filtered by power query, i would imagine that the file size will remain unchanged.