Azure Synapse Link for Cosmos DB

Azure Synapse Link for Azure Cosmos DB provides seamless integration between Synapse Analytics and Cosmos DB. As we discussed in the previous post, the analytics workloads have evolved from the traditional structured data warehouse requiring ETL (Extract Transform Load) integration to a more flexible and semi-structured big data and multi format data analysis. Azure Synapse Link caters to this by providing a hybrid transactional and analytical processing (HTAP) capability within the Azure cloud.

                                           Fig. Azure Synapse Link

As can be seen in the above diagram, the Azure Cosmos DB provides a built-in Analytical column store, which connects seamlessly with Azure Synapse Link. This connection provides data engineers and data scientists with the ability to process and analyse transactional data in near real-time latency using Synapse SQL or Synapse Spark.

The key advantage of using Azure Synapse Link is that it provides seamless integration without the need for ETL pipelines. Therefore, this approach is called No ETL integration. The changes in the transactional data store are synced automatically to the Analytical Store within Cosmos DB. These changes then reflect instantly in Azure Synapse Analytics query results without the need for processing an ETL pipeline.

Since the Analytical and Transactional stores in Cosmos DB are provisioned independent of each other, there is no performance impact on the transactional workloads within Cosmos DB while querying the analytical store using Azure Synapse Link.

Reference: https://docs.microsoft.com/en-us/azure/cosmos-db/synapse-link?toc=/azure/synapse-analytics/toc.json&bc=/azure/synapse-analytics/breadcrumb/toc.json

One thought on “Azure Synapse Link for Cosmos DB

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: