[ON-SITE] Introduction and Coaching with POP Methodology and Tools

Annotation

Training is offered with a variety of portable, open-source performance tools and associated analysis methodology of the Performance Optimisation & Productivity Centre of Excellence (POP CoE). The tools are particularly suited for HPC applications using MPI message-passing, multi-threading and/or kernel offloading to GPUs, covering parallel execution profiling, tracing and energy-efficiency analyses.  Platform-agnostic execution efficiency and scaling metrics characterise performance, leading to in-depth specific analyses. A short introductory section will present the POP methodology and tools, followed by individual/team coaching to apply one or more tools to participants’ own application codes.

Benefits for the attendees, what will they learn:

Attendees will learn how to characterise application execution, scaling and energy efficiencies, and how to apply portable open-source performance analysis tools to measure and analyse the execution performance of their own application codes to identify and quantify the most significant potential optimisation opportunities.

Prerequisites

Experience with building and running an HPC application code on Karolina (or similar system); no experience is required with any of the specific performance tools; however, prior experience with (parallel) performance analysis and tools will be valuable.

Technical requirements

Participants who have application codes running on Karolina (CPU or GPU partitions) or similar systems will be offered coaching to analyse their execution performance with the portable open-source tools developed and used by the POP Centre of Excellence. A short/small yet representative testcase should be prepared ready for execution.

Level

beginner, intermediate, advanced

Language

English

Tutors

Tobias Dollenbacher is a member of the RWTH Aachen team developing and providing training with the CPT performance analysis tool.

German Llort is part of the BSC team, developing and providing training with the Paraver/Extrae performance analysis tools.

Ondrej Vysocky is part of the IT4I team, developing and providing training with the MERIC/RADAR execution efficiency analysis tools.

Brian Wylie is part of the JSC team developing and providing training with the Scalasca/Score-P/CUBE performance analysis tools.

Acknowledgements

POP3 receives funding from the European High-Performance Computing Joint Undertaking (JU) under grant agreement No 101143931 (POP3). The JU receives support from the European Union’s Horizon Europe research and innovation programme and Spain, Germany, France, Portugal and the Czech Republic.

EuroCC 3 has received funding from the European High-Performance Computing Joint Undertaking (JU) under Grant Agreement No. 101306701. The JU receives support from European Union‘s Digital Europe Programme and Germany, Albania, Austria, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Montenegro, Netherlands, North Macedonia, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Türkiye, Kosovo. The EuroCC 3 project received funding from the Ministry of Education, Youth and Sports of the Czech Republic.

This course was supported by the Ministry of Education, Youth and Sports of the Czech Republic through the e-INFRA CZ (ID:90254).

All presentations and educational materials of this course are provided under the Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) license.