Part 4 covers advanced topics. We start by discussing the advantages of using Graphical processing units (GPUs) to process big data. It turns out that the computing model of GPUs is very amenable to processing large datasets, especially N-dimensional arrays. We conclude this section by introducing Dask: a Python-based framework to do parallel processing across many computers, allowing us to scale out to many machines when we need to process very large amounts of data with complex algorithms.