A genomic device sends 8 KB of data every second to a data platform. Requirements: near-real-time analytics, flexible/parallel/durable ingestion, and delivering processed results into a data warehouse. Which design should a solutions architect choose?

Kinesis Data Streams is ideal for near-real-time data ingestion and processing, offering durability and parallel processing capabilities necessary for high-volume genomic data. Kinesis clients can perform the required near-real-time analytics directly on the stream. Amazon Redshift is a suitable data warehouse for analytical workloads, and Amazon EMR can efficiently load processed data into Redshift. Incorrect options: Kinesis Data Firehose is primarily for loading data into data stores, not for real-time analytics with Kinesis clients. Amazon RDS is a transactional database, not optimized for large-scale analytical data warehousing. Storing data in S3 first and using SQS/Kinesis for processing introduces latency, making it less suitable for near-real-time requirements. API Gateway and SQS are good for event-driven architectures but might introduce higher latency and overhead for continuous, high-volume streaming data compared to Kinesis Data Streams. Lambda is suitable for event-driven processing but may not scale as cost-effectively for continuous, high-throughput stream analytics as Kinesis clients.