A company has an on-premises monitoring solution that uses a PostgreSQL database for event persistence, but the database cannot scale for heavy ingestion and runs out of storage. A VPN to AWS is already in place. The solution must use managed AWS services to reduce ops overhead; provide an automatically scaling buffer with no admin; enable near-real-time dashboards; and support semi-structured JSON with dynamic schemas. Which combination of AWS components will satisfy these requirements? (Choose two.)

The correct options address all requirements. Kinesis Data Firehose provides an automatically scaling buffer with no administration, handling heavy ingestion and acting as a managed service. It can deliver data to Amazon ES. Lambda functions can process and transform semi-structured JSON with dynamic schemas before delivery. Amazon Elasticsearch Service (Amazon ES) is ideal for near-real-time dashboards and analytics on semi-structured data, and its integrated Kibana endpoint provides visualization capabilities. Incorrect options: Kinesis Data Streams require capacity provisioning, increasing operational overhead compared to Firehose. Amazon Aurora PostgreSQL is a relational database and struggles with semi-structured JSON and dynamic schemas, and while scalable, it doesn't offer the same near-real-time analytics and visualization capabilities for this data type as Amazon ES. Amazon Neptune is a graph database, unsuitable for general-purpose event persistence and near-real-time dashboarding of semi-structured JSON.