Which file format is used to store per-customer data in the Parquet folder structure described?

Which file format is used to store per-customer data in the Parquet folder structure described?

• A. CSV
• B. Parquet
• C. JSON
• D. ORC

Answer: (B)

Parquet is a columnar storage format designed for analytics, which makes it ideal for per-customer data organized in a folder structure. Because data is stored column-by-column, queries that only need a subset of fields read far less data, speeding up analytics and reducing I/O. Parquet files are self-describing, meaning they carry the schema, and they compress well, which is beneficial when you have many small per-customer files. The folder structure can be partitioned by customer (and often by date or other dims), and Parquet supports efficient pruning of irrelevant partitions, so you scan only the data you need for a given customer. While formats like CSV or JSON are plain text and less efficient for large-scale analytics, and ORC is also columnar, the described Parquet folder layout indicates Parquet is the chosen format for storing per-customer data.

Parquet is a columnar storage format designed for analytics, which makes it ideal for per-customer data organized in a folder structure. Because data is stored column-by-column, queries that only need a subset of fields read far less data, speeding up analytics and reducing I/O. Parquet files are self-describing, meaning they carry the schema, and they compress well, which is beneficial when you have many small per-customer files. The folder structure can be partitioned by customer (and often by date or other dims), and Parquet supports efficient pruning of irrelevant partitions, so you scan only the data you need for a given customer. While formats like CSV or JSON are plain text and less efficient for large-scale analytics, and ORC is also columnar, the described Parquet folder layout indicates Parquet is the chosen format for storing per-customer data.