Exam PL-300 topic 2 question 4 discussion

A, only records with state finished are used D, personally I would transform the column to a date format and not split it since only the date part is used Not E, All the cancelled rows are already deleted with A and when a order is not cancelled it will contain a null value

It says: You need to reduce the LOAD times of the query without affecting the analysis. Only answers A and E can reduce the load times. D may reduce only the time needed to process the data. Someone said that E is not the answer because: "All the cancelled rows are already deleted with A and when a order is not cancelled it will contain a null value". You must read again the decription cause it says " Each answer presents a COMPLETE solutiuon" not part of a solution.

A (eliminar las filas con "Cancelado" en el estado) y E (eliminar las fechas de cancelación) es adecuada, ya que elimina la información innecesaria

AD are the correct answers

The filter applied in 'A' excludes all rows for 'Canceled', therefore a column for 'Canceled Date' would only contain NULL values and become useless

I selected E because in the question it says each correct answer presents a complete solution

Redueses the load and nowhere it mentioned to remove the column.

The correct answers are A and E. The reason why the correct answer is not D is because that would not reduce the load time as required in the question.

E can't be the answer as it removes the Canceled_Date column but not the rows

They question is asking to reduce load times. Automatically you should be looking at cutting off unnecessary columns. Only Sales with Status of FINISHED is needed, so you remove all Canceled rows. So thats A as one COMPLETE solution. Common sense should tell you since we dont need any Sale that is canceled, CANCELED DATE column MUST ALSO GO! D will not help performance (load times), it will only help the analysis. We were asked to improve the performance

A. Remove the rows in which Sales[Status] has a value of Canceled: Removing rows that are not required for analysis (i.e., where Sales[Status] is 'Canceled') will reduce the overall size of the dataset, making queries faster. Since only rows with a Status of 'Finished' are relevant for analysis, removing rows with 'Canceled' will help in reducing the load time. D. Split Sales[Sales_Date] into separate date and time columns: If you're only using the date part of the Sales_Date field, splitting it into separate date and time columns can be beneficial. By storing the date and time separately, Power BI can focus on just the date portion, reducing unnecessary data processing for queries that only require the date part, which improves performance.

AD not E

I understand removing rows in Sales[Status] = "Cancelled" will also remove rows in Sales[Canceled_date] reporting any date (will remain only null values in this column) but to "Reduce the load times" we should decrease the size of the data model so I would also vote for the entire column Sales[Canceled_date]. Splitting the Sales[Sale_date] would increase performance of the query as the "date" part will be isolated without the "time" details, but this would increase also the size as it would add an additional column to our model (from "sale_date" single column to "sale_date"+"sale _time")

There is no cancel date in Sales column, just a text value (status)

Removing rows will increase query performance and Splitting the Sales_Date column make comparison faster.

Said AD fisrt. But I changed my mind. D improves the analysis but doesn't change the performance, while D does.

I don't agree with the community in my opinion A and D are the correct ones.