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.

• ✓

  Pagination using LastEvaluatedKey until no further key is returned

  Why this is correct

  DynamoDB `Scan` and `Query` operations return results in 1MB chunks. To retrieve thousands of items, the developer must repeatedly call the API, passing the `LastEvaluatedKey` from the previous response as the `ExclusiveStartKey` in the subsequent request. This pagination continues until `LastEvaluatedKey` is no longer present in the response, indicating all items matching the criteria have been retrieved. This is the standard and most efficient way to handle large result sets.

• ✗

  A larger Lambda memory setting only

  Why it's wrong here

  Increasing a Lambda function's memory setting primarily allocates more CPU and network bandwidth, which can speed up processing within the function. However, it does not alter the fundamental behavior of DynamoDB API calls, which are inherently limited to returning a maximum of 1MB of data per request for `Scan` or `Query` operations. Therefore, a larger memory setting alone will not enable the retrieval of thousands of items in a single API call.

• ✗

  A global secondary index with the same key

  Why it's wrong here

  A Global Secondary Index (GSI) is designed to support different query patterns by allowing non-primary key attributes to be used as the index key. Creating a GSI with the *same* key as the base table offers no new access patterns or performance benefits for the stated requirement, as it essentially duplicates the primary key structure. Furthermore, even queries against a GSI are subject to the same 1MB result set limit per API call, necessitating pagination for large datasets.

• ✗

  Strongly consistent reads on every request

  Why it's wrong here

  Strongly consistent reads ensure that the data returned is the most up-to-date version, reflecting all successful writes prior to the read. While crucial for applications requiring immediate data accuracy, choosing strongly consistent reads does not impact the pagination mechanism or the 1MB data limit imposed on DynamoDB `Scan` or `Query` operations. The developer would still need to paginate through results using `LastEvaluatedKey` regardless of the consistency model chosen.

Go deeper

Related to this question