Picture of Ludovic Capelli

Instructor: Ludovic Capelli
Edinburgh Parallel Computing Centre
University of Edinburgh
l.capelli@epcc.ed.ac.uk

Ludovic Capelli is a teaching fellow at EPCC, which he joined after doing research in the optimisation of programming models for graph processing. EPCC is based at the University of Edinburgh, Scotland, UK. Among other activities, EPCC hosts and supports the UK national supercomputer service ARCHER2.

Course Description

This track is a gentle introduction to shared memory parallel programming and GPU offloading, using OpenMP. This track does not assume existing knowledge in parallel programming, other than a familiarity to compiled programming languages.

The teaching material, which will be available in both C and FORTRAN, will cover the basic of OpenMP for CPUs, from basic parallelisation and work sharing, through load-balancing and synchronisations. Similarly, for the GPU side, this track will provide you with a first experience towards offloading your code to GPUs as well as highlighting the fundamental performance challenges you will face, along with approaches to address them.

Timetable

Monday 10th of July - slides
  • 14:00 until 15:30
  • break from 15:30 until 16:00
  • 16:00 until 17:00

Tuesday 11th of July - slides
  • 9:00 until 10:30
  • break from 10:30 until 11:00
  • 11:00 until 12:30

Last modified: Tuesday, July 11, 2023, 2:13 PM