Skip to content
English
Contact us
  • There are no suggestions because the search field is empty.

I2024.2.7 BI Features and Improvements - System-Generated Distinct Count Aggregate Improvements

Business Problem

Many use-cases require fast evaluation of Distinct Count Measures.  Due to their low re-use rate, User Defined Aggregates (UDA's) were the best option to accelerate Distinct Counts, but such an approach is very labor intensive.  Pre-existing support for System-Generated Distinct Count Aggregates was so conservative that system-generated Distinct Count Aggregates were rarely generated.

Solution:

  1. Relax the hierarchy rules that discouraged Distinct Aggregates at higher levels.
  1. Improved Aggregate generation from inbound queries that have eligible dimension constraints.
  1. Opt-In Enablement and Configuration at the Model Level

Summary

  1. System-Generated agg table generation and usage for near-root hierarchy levels.
  1. System-Generated agg table generation and usage for queries with filters.
    1. Re-aggregation of Distinct Count Aggregate data is prohibited because the results would be incorrect.  Query Cache can handle most of these use-cases.
    2. Filters that restrict to a single dimension member will use Distinct Aggregates.
  1. Model-Scoped, Opt-In Enablement.
  1. Availability: i2024.2.7 and c2024.11.0