Webinar: POP CoE performance metrics: a beginner’s guide with Score-P, Scalasca & CUBE
This webinar introduces the POP CoE and its services. It provides an overview of the fundamental POP metrics employed in the assessment of parallel efficiency and scalability. Through demonstrations using tools such as Score-P, Scalasca and CUBE, participants will learn how to compute and interpret these metrics in practice. This session is aimed at those who want to understand standardised approaches to evaluating and improving HPC application performance.
This webinar was held June 16, 2026 and it’s the third part of a Webinar Series on Profiling.
About the speakers:
Ilya Zhukov is a senior HPC consultant at the Jülich Supercomputing Centre, with over eight years’ experience supporting users of large-scale HPC systems. He has been closely involved in the POP CoE, SiVeGCS and EPICURE projects at JSC, providing hands-on support to researchers across diverse scientific domains. His primary focus is on helping users exploit HPC systems efficiently and improve the performance and scalability of their applications.
Marc Schlütter is a research scientist at the Jülich Supercomputing Centre since 2010 with a focus on performance tools development in the context of the instrumentation and measurement framework Score-P. As part of projects like Mont Blanc, DEEPSEA and POP his focus has been support of application developers through a combination of user support, training and performance analysis.
About the moderator:
António Sousa is an Associate Professor at the Department of Informatics of the University of Minho and a Senior Researcher at the High-Assurance Software Laboratory (HASLab) at INESC TEC, where he serves as Co-coordinator. His research interests focus on dependable distributed systems, with particular application to distributed database systems, large-scale distributed systems, cloud computing management, and high-performance computing — notably on topics related to the management and efficiency of HPC systems. He serves as Director of Operations of the EuroHPC Deucalion supercomputer, hosted by the National Center for Advanced Computing, and Portugal’s national HPC flagship infrastructure, playing a central role in its operation and strategic development. He has been involved in several European HPC projects, RISC2, EUROCC1 and 2 and currently in EPICURE and HANAMI.
