Skip to main content

Public preview: Azure Synapse Link for Azure Cosmos DB partitioning Spark 3.1

Published date: March 15, 2022

Partition your Azure Cosmos DB analytical data using keys that are critical for your business logic to achieve better query and data load performance. Partition pruning results in faster analytical queries and loads as less data is scanned in the partitioned stores. You can also benefit from the data processing improvements if your workloads have a significant number of updates or deletes. 

To get started, please refer to our documentation and learn more about Apache Spark 3.1 in Azure Synapse Analytics.

  • Azure Cosmos DB
  • Features

Related Products