Arcadia Data's proprietary NDVP() function provides greater accuracy for approximate distinct count on very large data samples.

The name of this function is an acronym for "NDV + Precision". This aggregate function computes a more precise approximate distinct count than the similar Impala NDV Function. By default, NDVP() has a precision value of 14 compared to NDV() precision value of 10.

This function uses more memory, and has slightly increased processing times. Analytical views that contain NDVP() aggregates use more storage space compared to analytical views that use the NDV() function.



Returns: Double