A data scientist has developed a feature engineering notebook using the pandas library. However, as the data volume increases, the notebook's runtime escalates and processing speed drops in proportion to the data size. Which tool should the data scientist consider to efficiently scale their notebook for big data with minimal refactoring?