A DynamoDB table throttles during a promotion: reads hit a single 'promo' partition key, while the rest of the table sits idle. What's the fix?
A.Increase provisioned RCUs
More RCUs don't help — the per-partition limit is the bottleneck, not table capacity.
B.Write-shard the hot key — append a random suffix to spread the item across partitions
Correct — sharding the hot key is the documented hot-partition remedy.
C.Switch the table to on-demand mode
On-demand raises table capacity but the same per-partition ceiling still applies.
D.Add a global secondary index on the promo key
The GSI copies the same hot key — the problem follows the data.
Why this answer
Hot partitions cap throughput regardless of table-level capacity — a single partition supports ~1,000 WCU/3,000 RCU. Write-sharding (key + random suffix) spreads the hot item's traffic across partitions; a GSI can't fix a hot partition key either.
Correct answer: B · Last reviewed 2026-11-01
Official AWS references
Practice this interactively
Answer without seeing the solution first in the Database Selection & Scaling practice session.
Practice Database Selection & Scaling