You work for a large ecommerce company. You store your customer's order data in Bigtable. You have a garbage collection policy set to delete the data after 30 days and the number of versions is set…

Question

Options

• ASet the expiring values of the column families to 29 days and keep the number of versions to 1.
• BUse a timestamp range filter in the query to fetch the customer's data for a specific range.
• CSchedule a job daily to scan the data in the table and delete data older than 30 days.
• DSet the expiring values of the column families to 30 days and set the number of versions to 2.

• A
  19% (5)
• B
  70% (19)
• C
  7% (2)
• D
  4% (1)

Explanation

Option B is correct because Bigtable's garbage collection is not instantaneous - it runs during background compactions, meaning data past the 30-day threshold can still be physically present and queryable until the next compaction cycle. Adding a timestamp range filter to the query enforces the 30-day window at read time, giving analysts a precise and immediate boundary without waiting for GC to catch up, and it requires no infrastructure changes or additional cost.

A is wrong because reducing expiry to 29 days doesn't fix the root cause - GC still runs lazily, so stale data can still appear; it also unnecessarily tightens the retention window. C is wrong because a daily scan-and-delete job adds significant operational overhead and cost, directly violating the constraint. D is wrong because increasing versions to 2 would actually surface more old data to analysts, making the problem worse, not better.

Memory tip: Think of Bigtable GC as "eventual" - data will be deleted, just not right now. When you need a hard time boundary at query time, a timestamp filter is your scalpel; GC policy is just the scheduled cleanup crew.

No community discussion yet for this question.