Your teammate has asked you to review the code below. Its purpose is to efficiently add a large number of small rows to a BigQuery table. Which improvement should you suggest your teammate make?

Question

Exhibit

Options

• AInclude multiple rows with each request.
• BPerform the inserts in parallel by creating multiple threads.
• CWrite each row to a Cloud Storage object, then load into BigQuery.
• DWrite each row to a Cloud Storage object in parallel, then load into BigQuery.

• A
  94% (50)
• B
  4% (2)
• D
  2% (1)

Explanation

BigQuery's streaming insert API (tabledata.insertAll) accepts an array of rows per request, so inserting one row per API call for a large number of small rows is highly inefficient - it wastes API quota, increases latency due to per-request overhead, and can hit rate limits quickly. The most impactful improvement is to batch multiple rows into a single API request (Option A), which dramatically reduces the number of API calls and improves throughput. Option B (parallel threads) still sends individual rows and doesn't address the per-row API overhead. Options C and D (write to Cloud Storage then batch load) would work but introduce unnecessary complexity and latency for streaming inserts; the question specifically asks for an improvement to the existing row-insertion approach.

No community discussion yet for this question.