Your retail app writes heavy OLTP traffic to Azure Cosmos DB. The analytics team wants near-real-time dashboards and ad hoc queries without impacting transactional throughput or building ETL pipelines. What should you do to enable this HTAP scenario?

The correct options both describe Azure Synapse Link for Cosmos DB, which is designed for Hybrid Transactional/Analytical Processing (HTAP) scenarios. Enabling the analytical store on Cosmos DB containers and connecting an Azure Synapse workspace via Synapse Link allows for near-real-time analytics on operational data without impacting transactional performance. The analytical store is a fully isolated column store that automatically syncs data from the transactional store. Querying this analytical store with Synapse Spark or serverless SQL pools avoids consuming Request Units (RUs) from the transactional store, ensuring no impact on OLTP workloads. Querying the transactional store directly from Synapse serverless SQL pool would consume RUs and negatively impact transactional throughput, which is contrary to the requirement. Exporting data via change feed to Azure Data Lake Storage and then ingesting it into Synapse tables involves building ETL pipelines, which the scenario explicitly aims to avoid.