Explanation:
To effectively reduce BigQuery costs, it's crucial to avoid scanning excessive data. Using SELECT * can lead to scanning more data than necessary, thus increasing costs. Clustered tables help in organizing data more efficiently, reducing the amount of data scanned during queries. The bq --dry_run command is recommended for estimating query costs, not --estimate_bytes. Partitioning and clustering are both beneficial for limiting scanned data and reducing costs. However, using LIMIT on non-clustered tables does not reduce the amount of data scanned. For more details, refer to BigQuery best practices for performance.
Ultimate access to all questions.
No comments yet.
To optimize BigQuery usage and reduce costs for data analysts, which two practices should you recommend?
A
Utilize the bq --dry_run command to estimate query costs before execution
B
Implement clustered tables to minimize data scanned
C
Refrain from using SELECT * to avoid scanning unnecessary data
D
Limit the use of partitioned tables to reduce complexity
E
Apply LIMIT clauses exclusively to non-clustered tables for efficiency