A Kinesis data stream has highly variable, unpredictable traffic — sometimes near zero, sometimes bursting. Which capacity mode avoids over/under-provisioning?
A.Provisioned shards — fixed cost predictability
Provisioned wastes money on idle capacity during troughs.
B.On-demand mode — automatic shard scaling
Auto-scaling capacity = on-demand. Correct.
C.Kinesis Firehose instead of Data Streams
Firehose is a different delivery service — not a capacity answer.
D.SQS standard queue
SQS is message queuing — different ordering/consumer model.
Why this answer
On-demand mode scales shard capacity automatically (4MB/s in, up to 200MB/s out baseline expanding as needed) — best for variable/unpredictable loads, ~40% cheaper for spiky traffic. Provisioned mode is for predictable high-volume streams.
Correct answer: B · Last reviewed 2026-11-02
Official AWS references
Practice this interactively
Answer without seeing the solution first in the Scalability & Elasticity practice session.
Practice Scalability & Elasticity