Forum Discussion
Calc table Performance issues in composite models
The single most important factor is the cardinality of the join columns. Examine some of your measures in DAX Studio and you will find that the queries include enumerations of the values in the join column. If the cardinality is more than a couple thousand then you very quickly end up in situations where the query is slow not because it is complex, but because it is big, physically big. You can have scenarios where the query text is multiple gigabytes in size, and these gigabytes have to travel through the network for every single query.
Keep the link cardinality as low as possible.
The single most important factor is the cardinality of the join columns. Examine some of your measures in DAX Studio and you will find that the queries include enumerations of the values in the join column. If the cardinality is more than a couple thousand then you very quickly end up in situations where the query is slow not because it is complex, but because it is big, physically big. You can have scenarios where the query text is multiple gigabytes in size, and these gigabytes have to travel through the network for every single query.
Keep the link cardinality as low as possible.
As I continue to learn, I suspect my cardinality issue is attempting to do this with a "day' calendar column. When I flipped to a 'year' column, things started working much better.
I do think there are times when the service is more efficient than normal. I know, late Saturday afternoon - things were working great, and then Sunday morning - the old memory errors seemed to pop up.
Thanks for the input, I am slowly learning!
