Parallel programming with PyCOMPSs

PyCOMPSs is a parallel programming model that aims to ease the development of applications for distributed computing platforms (i.e, HPC clusters, clouds, or container-managed infrastructures).

This poster will give an overview of PyCOMPSs, including new features, such as: the ability of reading/writing from streamed data, failure management at task level, and integration with NUMBA.

The poster will also include a section for the dislib, a machine learning library parallelized with PyCOMPSs.

This poster is divided in the following sections:

logo1 logo4 logo6

logo5 logo2 logo8

logo7 logo3