This is a short course for the curriculum of Mathematics and Statistics for Life and Social Sciences

Laura del Río Martín

Laura del Río Martín
(University of Trento)
Personal website
Bio Laura del Rio received her PhD in Applied Mathematics from the University of A Coruña (Spain) in 2020. She is currently an assistant professor at the University of Trento. Her current research interests include numerical analysis and numerical simulations applied to multiphase compressible flows, solid mechanics and acoustics. She has in-depth knowledge in high-performance computing and massively parallel simulations.

Prerequisites

  • A basic knowledge of Python is required
  • Python will be used in all examples and practical sessions

Course Description

Numerical modeling and simulation techniques have become essential to address the new challenges of today’s society, thanks to new ICT technologies. In recent decades, there has been a significant advancement in the development of techniques for solving real-world problems, such as modeling flows around turbine blades, studying the mechanical and vibrational behavior of vehicles traveling at high speed on a road, modeling weather predictions, or simulating blood flow in arteries and veins. One of the biggest challenges for the scientific community is to deal with these complex multiphysics phenomena and achieve high accuracy while reducing the computational cost of the simulation. For this reason, tackle problems in real conditions, implementing efficient codes capable of using High-Performance Computing (HPC) technologies and infrastructures is essential. By doing so, simulations with very complex 3D geometries and sophisticated physical phenomena may be carried out, reducing their runtime.

The main objectives of this seminar are to introduce the participants to the paradigm of parallel computing. They will learn the basic concepts of parallelization and how to perform an efficiency analysis to take full advantage of parallel computing. In addition, they will learn how to use the high-performance infrastructures available at the university. The seminar will be hands-on so that participants will use all the tools presented in the course.

List of topics

  • Introduction to Parallel Programming: comparative of sequential and parallel computation and explanation of the parallel architectures
  • Description of the cluster of the University of Trento: hardware, software, job processing
  • Introduction to MPI and OpenMP
  • Performance and scalability analysis
  • Hands-on sessions on parallel computing

Schedule

  • 2023/06/07 10.30-13.30, Room A201 @ Povo 1
  • 2023/06/09 10.30-13.30, Room A202 @ Povo 1

Details

  • Venue: Povo1, room A201
  • Language: English
  • The participation is free. Please fill this form to confirm your participation. Deadline for registration: June 5th at 11:59 am
  • For further information, please contact Prof. Claudio Agostinelli

Material (Restricted access, user: HPC2023)

User Guide

Scripts

Codes

20230607

20230609