How do you optimize a slow-running SQL query that joins a massive transaction history table with a supplier risk profile table?
