Performance Tuning and Troubleshooting Tabular Models - SQL Server Analysis Services 2012 Cube Development - page 400

You have added over one million rows to the sheet, but you only have room
for about two-thirds of the available data.
11. In the field list, expand the Calendar hierarchy in the Date table, right-click
on the CalendarYear field, and add this field as a slicer.
Let's stop here because this was not a good user experience, and we haven't
achieved our objective. Many users would be frustrated and either call the help desk
or just give up at this point.
There's more…
The solution to this problem is actually quite simple but requires that users are edu-
cated about the practicality of consuming any significant volume of data.
If the same level of detailed reporting is a requirement, filters should be applied be-
fore selecting the fields that would result in cross-joining values on the rows. In our
case, we can filter by year and month to reduce the result size and improve per-
formance. Repeat the previous exercise, but prior to step 7 of the previous section,
perform the following additional steps:
1. Add CalendarYear and MonthNumberOfYear as filters.
2. Select a value from each of these cells before adding other fields to the rows
or columns of the PivotTable report.
Another approach is to help users understand the value of summarizing and slicing
data, rather than drilling down to raw details in a single report. If slicers or filters were
applied on dimensional members such as product category, product, and customer,
we could have displayed daily details without adding the product and customer to the
details of the report and creating the cross-join problem. Multiple reports can also be
created to get customer details in one list and product details in another list, with in-
teractive filters and slicers that can bring us to the same information in a simple and
more effective presentation.