How to remove data efficiently

[Anonymous]

Dear Community,

I just started with Power BI- Query and I do have some troubles by removing columns and unnecessary data.

My question: What is the best way to remove data when it comes to data efficiency?

1) via removing the column?

2) by “Choose columns” and only select the columns i may need ?

3) maybe with a Measure (if possible?)?

I am trying to make the report as efficient as possible.

Thank you very much in advance.

[Vijay_A_Verma · Accepted solution]

Whether you use Select columns or remove columns, it will use Table.SelectColumn function only. So, both are same as they are executing the same statements. So 1 and 2 are same, no difference.

DAX will not remove a column. Measures can produce final result values. 

In general, measures are faster than PQ (exceptions aside). But in case of calculated columns, PQ is faster (exceptions aside).

In Power BI, you need to reduce the size of rectangle (rows * columns) as much as possible. Hence, before doing any transformations or calculations, remove as many columns as possible to reduce the size of rectangle. You should also filter out as many rows as possible.

Then you can decide whether to use DAX or PQ as per your need. 

[Anonymous]

Dear @Anonymous , @Vijay_A_Verma,

Thanks for your reply and support 🙂

Best regards,

[Anonymous · Accepted solution]

Normally, i use 2) by “Choose columns” and only select the columns i may need ? because in this approach, it is easy and quick for you to choose columns to keep when you have so many columns. In this option, the top serach bar also can help you find the columns fast. 

[Vijay_A_Verma · Accepted solution]

Whether you use Select columns or remove columns, it will use Table.SelectColumn function only. So, both are same as they are executing the same statements. So 1 and 2 are same, no difference.

DAX will not remove a column. Measures can produce final result values. 

In general, measures are faster than PQ (exceptions aside). But in case of calculated columns, PQ is faster (exceptions aside).

In Power BI, you need to reduce the size of rectangle (rows * columns) as much as possible. Hence, before doing any transformations or calculations, remove as many columns as possible to reduce the size of rectangle. You should also filter out as many rows as possible.

Then you can decide whether to use DAX or PQ as per your need.