How do you handle memory management when processing a dataset that is significantly larger than your available RAM?
