Skip to content

Schedule

Names of lecturers are preliminary and may still change.

DAY 1 - Tuesday 11/06/2024
09:00 EEST
08:00 CEST 		Welcome and Introduction
Presenters: Anne Vomm (LUST) and Harvey Richardson (HPE)
09:10 EEST
08:10 CEST 		LUMI Architecture, Programming and Execution Environment
Presenters: Harvey Richardson (HPE)
  • Recap of the LUMI architecture, programming, and execution environment.
09:50 EEST
08:50 CEST 		Exercises (session 1)
10:10 EEST
09:10 CEST 		Compilers, libraries, runtime, single-core optimisations
Presenter: Alfio Lazzaro (HPE)
  • Topics relevant to performance and optimization.
11:10 EEST
10:10 CEST 		Break (20 minutes)
11:30 EEST
10:30 CEST 		Introduction to Performance Analysis with Perftools
Presenter: Thierry Braconnier (HPE)
  • Overview of the Cray Performance and Analysis toolkit for profiling applications.
  • Demo: Visualization of performance data with Apprentice2/3.
12:10 EEST
11:10 CEST 		Exercises (session 2)
12:30 EEST
11:30 CEST 		Lunch break (60 minutes)
13:30 EEST
12:30 CEST 		Advanced Performance Analysis - Demo
Presenters: Alfio Lazzaro (HPE)
  • Automatic performance analysis and loop work estimated with perftools.
  • Communication Imbalance, Hardware Counters, Perftools API, OpenMP.
  • Compiler feedback.
14:00 EEST
13:00 CEST 		Exercises (session 3)
14:30 EEST
15:30 CEST 		Break (30 minutes)
15:00 EEST
14:00 CEST 		Exercises (session 4)
17:00 EEST
16:00 CEST 		Open Questions & Answers
Participants are encouraged to continue with the exercises in case there should be no questions.
17:30 EEST
16:30 CEST 		End of the workshop day
DAY 2 - Wednesday 12/06/2024
09:00 EEST
08:00 CEST 		MPI Topics on the HPE Cray EX Supercomputer
Presenter: Harvey Richardson (HPE)
  • High level overview of Cray MPI on Slingshot.
  • Useful environment variable controls.
  • Rank reordering and MPMD application launch.
09:50 EEST
08:50 CEST 		Exercises (session 5)
10:20 EEST
09:20 CEST 		Break (30 minutes)
10:50 EEST
09:50 CEST 		IO Optimizations
Presenters: Harvey Richardson (HPE)
  • Tips for optimising parallel bandwidth for a variety of parallel I/O schemes.
  • Introduction to the Lustre filesystem, controls and being nice to Lustre.
  • Advanced Parallel I/O considerations: Further considerations of parallel I/O and other APIs.
11:30 EEST
10:30 CEST 		Exercises (session 6)
12:00 EEST
11:00 CEST 		Lunch break (60 minutes)
13:00 EEST
12:00 CEST 		AMD Profiling Tools and GPU optimisations 1
Presenter: Samuel Antão (AMD)
14:00 EEST
13:00 CEST 		Exercises (session 7)
14:30 EEST
13:30 CEST 		Break (20 minutes)
14:50 EEST
13:50 CEST 		AMD Profiling Tools and GPU optimisations 2
Presenters: Samuel Antão (AMD)
15:40 EEST
14:40 CEST 		Exercises (session 8)
17:00 EEST
16:00 CEST 		Open Questions & Answers
Participants are encouraged to continue with the exercises in case there should be no questions.
17:30 EEST
16:30 CEST 		End of the workshop day
DAY 13 - Friday 24/10/2025
09:00 EEST
08:00 CEST 		Best practices: GPU Optimization, tips & tricks / demo
Presenter: Samuel Antão (AMD)
10:30 EEST
09:30 CEST 		Break (30 minutes)
11:00 EEST
10:00 CEST 		Exercises (session 9)
12:00 EEST
11:00 CEST 		Open Questions & Answers
Participants are encouraged to continue with the exercises in case there should be no questions.
12:30 EEST
11:30 CEST 		End of the workshop