A cybersecurity company is collecting on-premises server logs, mobile app logs, and IoT sensor data. The company backs up the ingested data in an Amazon S3 bucket and sends the ingested data to…

While creating two Kinesis Data Firehose streams is serverless and scalable, this approach lacks the explicit buffering and robust handling of severe 'sudden changes in data flow' that a dedicated Kinesis Data Stream provides, potentially leading to throttling or backpressure issues for producers.

This option describes a robust and scalable solution, similar to the intended solution for D, but if the truncated portion of D implies a configuration with fewer Firehose streams (e.g., direct OpenSearch ingestion from KDS), then B might be considered less cost-effective due to the explicit use of two Firehose streams for both destinations.

Creating only one Amazon Data Firehose delivery stream to OpenSearch Service fails to meet the requirement of backing up the ingested data to an Amazon S3 bucket.

Creating one Amazon Kinesis Data Stream centralizes the ingestion pipeline, providing a scalable and durable buffer to absorb sudden changes in data flow from diverse sources. One or more Amazon Kinesis Data Firehose delivery streams can then consume from this Kinesis Data Stream to cost-effectively deliver the data in a serverless manner to both Amazon S3 for backup and Amazon OpenSearch Service for analysis, meeting all scalability and serverless requirements.