A company ingests a large set of clickstream data in nested JSON format from different sources and stores it in Amazon S3. Data Analysts need to analyze this data in combination with data stored in…

Question

Options

• AUse Apache Spark SQL on Amazon EMR to convert the clickstream data to a tabular format. Use
• BUse AWS Lambda to convert the data to a tabular format and write it to Amazon S3. Use the
• CUse the Relationalize class in an AWS Glue ETL job to transform the data and write the data back
• DUse the Amazon Redshift COPY command to move the clickstream data directly into new tables

• A
  14% (3)
• B
  5% (1)
• C
  77% (17)
• D
  5% (1)

Explanation

Nested JSON is inherently non-relational and cannot be directly queried by Amazon Redshift without transformation. AWS Glue provides the Relationalize class specifically designed to flatten nested and complex JSON structures into relational (tabular) tables, handling arrays by creating separate tables with foreign key relationships. An AWS Glue ETL job using Relationalize can: (1) read nested JSON from S3, (2) automatically flatten it into relational tables, (3) write the output back to S3 in a Redshift-compatible format (e.g., Parquet), and (4) be scheduled for automation. Redshift can then query this data via Redshift Spectrum or COPY. This is cost-effective (serverless ETL, no always-on infrastructure) and automated (Glue job triggers). Option A (EMR/Spark) works but adds operational overhead. Option B (Lambda) has runtime/memory limits ill-suited for terabyte-scale nested JSON. Option D (Redshift COPY) cannot natively handle deeply nested JSON structures.

No community discussion yet for this question.