Be aware, that when creating User Dimensions with multiple members, each of these members will typically send a data request to the data warehouse according to the criteria settings of the member.
In cases with many members (e.g., 20 or more), or even with less members in combination with multiple other standard dimensions, you may see a considerable performance decrease. The time to load the document and the stress applied on the server may grow considerably.
Standard dimensions, directly from the data warehouse, will usually perform better than User Dimensions. So, if you experience performance issues with User Dimensions, you may consider using an alternative standard dimension from the data warehouse, if available. If not available, you may consider creating that standard dimension in the data warehouse.
Example
Consider a scenario with these two tables.
The first table has been made with a Month dimension directly from the Data Warehouse. The second table contains a User Dimension with 12 members. The only purpose of the User Dimension is to display the shorter month names.
However, since the User Dimension contains 12 members, the second crosstab will be at least 12 times slower than the first crosstab when requesting data.
In this example, performance wise, it would be better to create a new attribute/dimension in the data warehouse, containing the shorter month names.
Comments
Please sign in to leave a comment.