How would you handle memory constraints when loading and processing a multi-gigabyte CSV file containing market data using Pandas?
