I agree, we're taking this work load over. right now it's a backup of production and it hasn't fully moved over yet. Trying to figure out the best ways to work with this pre-prod. It is mostly for history and queried against. Downtime wouldn't really be feasible long term. We can accept slower impact on nights or possibly a weekend time, but actual downtime is best to avoid if at all possible. It has 3.8 billion rows roughly partitioned across 20+ files.