Exam 70-778 topic 1 question 21 discussion

Actual exam question from Microsoft's 70-778

Question #: 21
Topic #: 1

You plan to create a Power BI report. You have the schema model shown in the exhibit. (Click the Exhibit button.)

The model has the following relationships:
✑ Store to District based on DistrictID
Sales to Store based on LocationID

✑ Sales to Date based on PeriodID
✑ Sales to Item based on ItemID
You configure row-level security (RLS) so that the district managers of the stores only see the sales from the stores they manage.
When the district managers view the Store by Items report, they see items for all the stores.
You need to ensure that the district managers can see items for the stores they manage only.
How should you configure the relationship from Sales to Item?

A. Select Assume Referential Integrity.
B. Change the Cardinality to One to Many (1:*).
C. Change the Cross filter direction to Both.
D. Change the Cardinality to One to one (1:1).

Show Suggested Answer

Suggested Answer: C 🗳️
References:
https://powerbi.microsoft.com/en-us/guided-learning/powerbi-admin-rls/

by Hien at Aug. 15, 2019, 3 a.m.

Comments

Submit Cancel

Hien

Highly Voted 5 years, 10 months ago

The reference changed to https://docs.microsoft.com/en-us/power-bi/service-admin-rls

upvoted 6 times

r8d1

5 years, 1 month ago

the relevant part from the above link "row-level security filtering uses single-directional filters, regardless of whether the relationships are set to single direction or bi-directional. You can manually enable bi-directional cross-filter with row-level security by selecting the relationship and checking the Apply security filter in both directions checkbox".

upvoted 1 times

...

AtmaNow

Most Recent 4 years, 5 months ago

Both ? - Is actually only for username() filtering.

upvoted 1 times

...

CDL

4 years, 8 months ago

2 key points here: Key-1, RLS setting, PBI desktop -> Manage Role -> add “DM” filter by DM. Link: https://www.youtube.com/watch?v=MxU_FYSSnYU Key-2, also 2 knowledge points (1) Cross filter function (read the example): https://docs.microsoft.com/en-us/dax/crossfilter-function (2) Many to many relationship (read the example: composite model): https://docs.microsoft.com/en-us/power-bi/transform-model/desktop-composite-models Ref Example Video: https://www.youtube.com/watch?v=44JKwL3Js10 My thought: Becoz there is no relationship between Store and Item. Store to Sales: 1 to many Item to Sales: 1 to many But we need Sum(Sales) meet both “Row context” at the same time, which means we need 2 different fields from 2 different table without any relationship, so that’s why we need “Cross Filter” to “cross” the “Sales” to establish “both” filters to satisfy measure Sum(sales) in “Sales”. So I go with C.

upvoted 3 times

...

JohnFan

5 years, 7 months ago

"When the district managers view the Store by Items report, they see items for all the stores." -- What causes this?

upvoted 1 times

Nishant

5 years, 6 months ago

Because the relationship b/w sales and item are single-sided and that is from Item to Sales direction, the filter does not reach the item table. If we make this relationship bi-directional the filter can reach to item from Store.

upvoted 15 times

kilowd

5 years, 4 months ago

For One-to-many relationships, the cross filter direction is always from the "one" side, and optionally from the "many" side (bi-directional)..Hope this helps

upvoted 2 times

...

Brunobsv

5 years, 7 months ago

For those wondering, it's C because by the relationship, many-to-one, sales table is unable to filter item, only item are able to filter table. That's my reading correct if I am wrong.

upvoted 2 times

...