A DynamoDB read path has become slow only during busy afternoons. Which CloudWatch metric should be checked first to see whether requests are being rejected for lack of read throughput?
Show answer and explanation
Answer: C
ReadThrottleEvents counts read requests that DynamoDB rejected because the table or an index exceeded its read capacity. A rising count during the slow afternoons points to a throughput limit, which can then be fixed with more capacity or a better key design. ItemCount and TableSizeBytes describe how much data is stored, and the number of indexes is a schema property. None of these three reveal rejected requests.
Why the other options are wrong
- A. ItemCount shows how many items the table holds. It says nothing about whether reads are being rejected during peak periods.
- B. TableSizeBytes reports storage consumed. A large table is not slow by size alone, so this metric does not indicate read throttling.
- D. The index count is a fixed schema detail visible in the table description, not a runtime metric that would show a slowdown during busy hours.