Skip to main content
Research and Business
A purple illustration of the planet with virtual connections

Performance Analysis Methods: Workshop Series:2025

 

Systematic performance analysis is a key activity in modern HPC code development. It shifts us from trial-and-error approaches toward an engineering-like process for delivering fast code. Skills to conduct performance analysis are essential for every modern HPC research software developer. Performance analysis is challenging: We must not only understand and utilise sophisticated analysis tools, but also follow a systematic workflow. In this workshop series, we will introduce key tools for analysing HPC code performance, but our primary focus will be on the workflow behind performance analysis:

  • Where do we begin?
  • What areas should we explore?
  • How can we transform ad-hoc performance analysis into a rigorous, hypothesis- and data-driven scientific workflow?

 

This workshop will equip participants with an understanding of performance analysis methodology based on the state-of-the-art in the field, along with practical skills in using analysis tools. Throughout the series, participants will collaborate on a joint report documenting findings for the broader community.

 

                students working toegther

 

 

Dates

  • 1 and 2 May: One day workshop (in person) in Durham
  • 6 May: Half day virtual workshop on intra-node, inter-node, I/O, node performance or GPU assessment
  • 8 May: Half day virtual workshop on intra-node, inter-node, I/O, node performance or GPU assessment
  • 13 May: Half day virtual workshop on intra-node, inter-node, I/O, node performance or GPU assessment
  • 15 May: Half day virtual workshop on intra-node, inter-node, I/O, node performance or GPU assessment
  • 20 May: Half day virtual workshop on intra-node, inter-node, I/O, node performance or GPU assessment
  • 20/21 May: One day workshop (in person) in Durham to wrap workshop series up

 

Requirements 

Participants have to have good knowledge in key HPC technologies. This means at least one compiled language (C++ or Fortran preferred), MPI and a multithreading technology (e.g. OpenMP). Knowledge in GPU programming is advantageous. No previous experience with performance analysis tools is required. Candidates should commit to participate in all sessions: to build a cohort, dropping in and out is problematic. We appreciate if candidates bring along one code that they plan to analyse throughout the workshop series with different tools and methodologies. However, we will also encourage participants to team up and look into codes unknown to them, i.e. there will be plenty of case studies on the table.

 

 

Register here 

 

 

For any other enquiries please contact idas.admin@durham.ac.uk