high-level parallel programming illustrated by Python and Julia examples

High-level parallel programming is important to decide whether a specific computation may benefit from parallelism. The rapid prototyping capabilities of a high-productivity language such as Python are illustrated with an example from numerical integration, using the vectorization provided by NumPy and SciPy. Julia is a language for numerical computing, aimed at performance, with a built-in LinearAlgebra module which gives direct access to the BLAS software library. Parallel computations with Julia are illustrated via a Jupyter notebook. While a formal definition of high-level parallel programming is hard to find, the important characteristics are that it is familiar (no new language needed), interactive (direct feedback), and personal (suitable on a personal computer). This allows to experience many concepts in supercomputing in a hands-on fashion.