E-MapReduce (EMR) Serverless Spark provides Spark sessions in workspaces for SQL development, notebook execution, and JDBC/ODBC connectivity.
Session types
EMR Serverless Spark supports the following session types:
-
SQL session: used for SQL development and queries. To run a query in the SQL editor, select an appropriate SQL session.
-
Notebook session: used to run notebooks. To develop a notebook, you must create or select an existing notebook session.
-
Spark Thrift Server: A Spark Thrift Server is an Apache Spark service that lets you connect to Spark and run SQL queries over Java Database Connectivity (JDBC) or Open Database Connectivity (ODBC). You can use it to integrate Spark with business intelligence (BI), data visualization, and other data analysis tools.