A DevOps Engineer Has a Single Amazon DynamoDB Table That

Multiple Choice

A DevOps Engineer has a single Amazon DynamoDB table that received shipping orders and tracks inventory. The Engineer has three AWS Lambda functions reading from a DymamoDB stream on that table. The Lambda functions perform various functions such as doing an item count, moving items to Amazon Kinesis Data Firehose, monitoring inventory levels, and creating vendor orders when parts are low. While reviewing logs, the Engineer notices the Lambda functions occasionally fail under increased load, receiving a stream throttling error. Which is the MOST cost-effective solution that requires the LEAST amount of operational management?

A) Use AWS Glue integration to ingest the DynamoDB stream, then migrate the Lambda code to an AWS Fargate task.
B) Use Amazon Kinesis streams instead of DynamoDB streams, then use Kinesis analytics to trigger the Lambda functions.
C) Create a fourth Lambda function and configure it to be the only Lambda reading from the stream. Then use this Lambda function to pass the payload to the other three Lambda functions.
D) Have the Lambda functions query the table directly and disable DynamoDB streams. Then have the Lambda functions query from a global secondary index.