Environment setup for Snowpark Connect for Spark¶
Before you can develop Spark workloads with Snowpark Connect for Spark, you need to set up your development environment and configure a connection to Snowflake.
You can develop Snowpark Connect for Spark workloads in the following environments:
Local IDE: Use your preferred local tools such as VS Code, Jupyter Notebooks, IntelliJ, or a terminal. The
snowpark-connectPython package is required for all languages. For Java and Scala projects, also add thesnowpark-connect-java-clientMaven dependency. These handle server lifecycle and session management automatically. You can also use open-source Spark libraries to connect directly to Snowflake’s hosted Spark Connect endpoint.Snowflake Workspaces: Use Snowflake Notebooks running on warehouses or in workspaces. The
snowpark-connectpackage is available as a notebook dependency.
For setup instructions, see the following topics: