Microsoft Azure Data Fundamentals DP-900 Practice Question

Apache Parquet is a columnar storage format that stores data by column rather than by row. Because it is column-oriented, analytical engines such as Synapse and Databricks can read only the columns they need, which reduces I/O. Parquet files also include metadata that describes the schema and apply efficient compression and encoding to reduce storage size. CSV, JSON, and fixed-width text are row-oriented formats that lack built-in column pruning and typically consume more space for large analytical datasets.