OSE code templates¶
We show how to parallelize a loop using the multiprocessing
and mpi4py
. The setup allows
to seamlessly switch between shared and distributed memory computing.
We collect resources and demonstrate parallelization with numba
. Our focus lies on the
analysis of nested parallelism and the working example is inspired by respy
.
We show how to set up a main-child application. We use the example of uncertainty propagation
using respy
as the motivating use-case.