Call a stored procedure in Redshift from Glue

This should be very similar to running any SQL command on Redshift through python. You could use a python-shell job which connects to Redshift to run a stored procedure. A python sample which shows how to connect to the Redshift cluster from python using psycopg2 library:

def loadfile(filename):
  conn_string = "dbname='dbname' port='port' user='user' password='password' host='host.eu-west-1.redshift.amazonaws.com'"
  conn = psycopg2.connect(conn_string)
  cursor = conn.cursor()
  cursor.execute("INSERT INTO tb_table (field1) VALUES ('field1')")
  conn.commit()
  cursor.close()
  return 

There are details on some more forums
https://www.quora.com/Can-I-connect-to-Redshift-using-Python
https://www.blendo.co/blog/access-your-data-in-amazon-redshift-and-postgresql-with-python-and-r/