If you want to work specifically with Pandas, use Pandaral·lel; for parallel distribution of work in Jupyter notebooks, use Ipyparallel. And, for more generic Python workloads, Dispy, Joblib ...