Assume that you are the only Azure Enterprise Data Analyst in your team. Your Project Manager has asked you to look at some techniques to improve data loading for relational data sources for your…

Question

Explanation

This question tests knowledge of performance optimization techniques for relational data sources in Power BI/Azure environments, specifically which methods genuinely reduce data load times versus those that are irrelevant or harmful.

Approach. The three correct techniques are A (Add Indexes), B (Design Distributed Tables), and E (Compressing Data). Adding indexes on source tables allows Power BI queries to retrieve rows faster by avoiding full table scans. Designing distributed tables (a core Azure Synapse Analytics concept using hash, round-robin, or replicated distribution) spreads data across compute nodes, dramatically improving parallel query throughput during large data loads. Compressing data reduces the volume of I/O between the source and Power BI, lowering network transfer time and storage pressure - both of which directly speed up loading. Option C (Materialize a date table) is a Power BI model best practice but addresses report calculation performance, not relational source loading. Option D (Truncating column values) discards data and offers no loading performance benefit.

Concept tested. Power BI performance optimization for relational data sources - specifically database-level techniques (indexing, distributed table design, compression) in Azure environments such as Azure Synapse Analytics and Azure SQL Database that reduce query execution time and data transfer overhead during the ETL/ELT load phase.

Reference. Microsoft Learn - 'Optimize Power BI performance' and 'Azure Synapse Analytics table distribution design guidance'

No community discussion yet for this question.