Skip to content

Accelerating Python with Multiple Nodes with MPI for Python

The Luxembourg SuperComputing Competence Center will host a half-day online course on accelerating Python using multiple nodes with MPI for Python. This course will introduce you to mpi4py and the core concepts of parallel computing across multiple nodes. You will gain hands-on experience in leveraging High-Performance Computing (HPC) to optimize Python programs, enabling them to scale efficiently across multiple processors and nodes.

For whom?

This workshop is ideal for those looking to enhance their Python-based HPC skills. Participants should have intermediate Python proficiency and a basic understanding of parallel programming (MPI). The session will provide a comprehensive introduction to utilizing HPC to optimize Python code for performance.

What will you learn and how?

In this workshop, you will learn how to harness the power of HPC clusters with mpi4py to optimize and accelerate your computational workflows. The course will guide you through configuring a SLURM environment, understanding the basics of mpi4py, implementing efficient data distribution techniques, and exploring practical applications in real-world scenarios.

Learning Outcomes

After completing this course, participants will be able to:

  • Configure a SLURM environment for running MPI-based applications.
  • Implement efficient data distribution using mpi4py.
  • Accelerate Python workflows by integrating MPI.

Prerequisites

  • Basic knowledge of SLURM.
  • A fundamental understanding of Python and NumPy.
  • Intermediate-level knowledge of parallel programming.

Compute Resources

Participants attending the training will receive access to the MeluXina supercomputer during the session. For more information about MeluXina, please refer to the MeluXina Overview and the MeluXina – Getting Started Guide.

Agenda

This half-day will be hosted online in Central European Time (CET) on May 7th, 2025 (9:00 AM – 12:30 PM CET).

Schedule:

  • Introduction to mpi4py
  • Configuring scripts to run with multiple MPI tasks
  • Practical use cases and applications

Important: Limited spots available (25 participants max)!

Contact people for more info:
Oscar CASTRO-LOPEZ, oscar.castro@uni.lu