I don’t really understand what query could cause a 40+Gb spike. There’s nothing in our system that could generate such a query 🤔