For a DirectQuery semantic model with a large dataset, which feature helps reduce query time for visuals on all pages?

For a DirectQuery semantic model with a large dataset, which feature helps reduce query time for visuals on all pages?

• A. User-defined aggregations
• B. Row-level security
• C. Columnstore indexing
• D. Automatic page refresh

Answer: (A)

User-defined aggregations let you create pre-aggregated tables at different grain levels within the DirectQuery model. Since DirectQuery keeps data in the source, visuals often issue queries that scan large base tables. By defining aggregated tables (for example, sums or counts by key dimensions) and mapping measures to those aggregates, the engine can answer many visuals from these smaller, pre-aggregated results rather than the full detail tables. This dramatically reduces the amount of data read and the complexity of joins, so visuals load faster and the improvement shows across pages because many visuals rely on the same aggregates. Row-level security affects who can see data, not how fast queries run across all pages. Automatic page refresh is about how often visuals update, not the query time itself. Columnstore indexing is a database-level optimization, which can help performance, but it’s not a model-level feature like user-defined aggregations that directly accelerates DirectQuery visuals across multiple pages.

User-defined aggregations let you create pre-aggregated tables at different grain levels within the DirectQuery model. Since DirectQuery keeps data in the source, visuals often issue queries that scan large base tables. By defining aggregated tables (for example, sums or counts by key dimensions) and mapping measures to those aggregates, the engine can answer many visuals from these smaller, pre-aggregated results rather than the full detail tables. This dramatically reduces the amount of data read and the complexity of joins, so visuals load faster and the improvement shows across pages because many visuals rely on the same aggregates.

Row-level security affects who can see data, not how fast queries run across all pages. Automatic page refresh is about how often visuals update, not the query time itself. Columnstore indexing is a database-level optimization, which can help performance, but it’s not a model-level feature like user-defined aggregations that directly accelerates DirectQuery visuals across multiple pages.