An organization is using Amazon DynamoDB with on-demand capacity. They notice that the 'ThrottledRequests' metric is non-zero during a specific hour each day. The table stores session data and has a partition key of 'user_id'. The workload is uniform. What is the most likely cause of throttling?
Trap 1: The table has on-demand capacity and cannot throttle.
This is incorrect. On-demand tables can throttle due to per-partition or account-level limits, so the statement 'cannot throttle' is false.
Trap 2: A global secondary index is consuming write capacity.
This is unlikely because the question does not mention a global secondary index (GSI). Even if a GSI exists, it would share the same on-demand capacity and not have separate limits that cause throttling unless its partition key is skewed, which is not indicated.
Trap 3: The partition key is not evenly distributed.
This is incorrect. The workload is explicitly stated as uniform, meaning the partition key 'user_id' is evenly distributed. Therefore, throttling due to an uneven distribution of the partition key is not the most likely cause.
- A
The table has on-demand capacity and cannot throttle.
Why wrong: This is incorrect. On-demand tables can throttle due to per-partition or account-level limits, so the statement 'cannot throttle' is false.
- B
The account-level read/write capacity limit is exceeded.
This is correct. Even with on-demand mode, there is an account-level throughput limit that, if exceeded, results in throttling. This is commonly encountered during periodic traffic spikes.
- C
A global secondary index is consuming write capacity.
Why wrong: This is unlikely because the question does not mention a global secondary index (GSI). Even if a GSI exists, it would share the same on-demand capacity and not have separate limits that cause throttling unless its partition key is skewed, which is not indicated.
- D
The partition key is not evenly distributed.
Why wrong: This is incorrect. The workload is explicitly stated as uniform, meaning the partition key 'user_id' is evenly distributed. Therefore, throttling due to an uneven distribution of the partition key is not the most likely cause.