DVA-C02 Development with AWS Services Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Decrease the batch size of the Kinesis event source mapping.

  Why it's wrong here

  Decreasing the batch size of the Kinesis event source mapping reduces the number of records processed by a single Lambda invocation. While this might slightly lower the write volume per invocation, it does not fundamentally address the underlying issue of DynamoDB's provisioned write capacity being exceeded over time. The total rate of events from Kinesis to Lambda, and thus the overall write attempts to DynamoDB, remains largely unchanged, making this an ineffective solution for persistent `ProvisionedThroughputExceededException` errors.

• ✓

  Implement retry logic with exponential backoff in the Lambda function.

  Why this is correct

  Implementing retry logic with exponential backoff in the Lambda function is the standard and most effective approach for handling `ProvisionedThroughputExceededException`. This exception indicates a temporary throttling by DynamoDB when its provisioned capacity is exceeded. Exponential backoff allows the Lambda function to automatically reattempt failed writes after increasing delays, giving DynamoDB time to recover capacity and successfully process the request, thereby smoothing out write spikes and preventing data loss.

• ✗

  Increase the number of shards in the Kinesis stream.

  Why it's wrong here

  Increasing the number of shards in the Kinesis stream enhances the stream's ability to ingest and process data, increasing its throughput capacity. However, this change directly impacts Kinesis's performance, not DynamoDB's write capacity units. If anything, a higher shard count might allow Kinesis to deliver events to Lambda at an even faster rate, potentially exacerbating the `ProvisionedThroughputExceededException` if the downstream DynamoDB table cannot keep up with the increased processing speed.

• ✗

  Increase the memory allocated to the Lambda function.

  Why it's wrong here

  Increasing the memory allocated to the Lambda function primarily boosts its computational resources, including CPU and available RAM, which can improve the performance of internal processing tasks. However, the `ProvisionedThroughputExceededException` is a server-side error originating from DynamoDB, indicating its write capacity has been exhausted. Lambda's memory allocation has no direct influence on the external DynamoDB table's provisioned throughput or its ability to accept more write requests.

Go deeper

Related to this question