You might take a look at this: How to allow Spark to access Microsoft SQL Server – Big Data Support
From a Tableau perspective, it would just be leveraging the "Spark SQL" connector in Desktop and entering in the appropriate credential and server info: Spark SQL
That said, Microsoft SQL Server has a native connector to Tableau; the ODBC connection tends to be fairly well optimized by my experience. Unless you need to go through the JDBC route for some reason, I always would recommend going through the native connector instead.
You would use the Spark Data source API to connect to SQL server.
In Spark - SQL
CREATE TABLE jdbcTable
USING <sql server jdbc driver>
Once you have defined this table in Spark you can access this from Tableau using the Spark thrift server connection.