Associate Professor, HPC. University of Basel. Switzerland.
“Performance Optimization & Portability: Pathways for the Era of Heterogeneous HPC Systems“
20 February 2025
Abstract
Contemporary High Performance Computing (HPC) is increasingly defined by heterogeneity, characterized by a diverse array of devices and a multitude of cores per device within each node. Looking ahead, the HPC landscape is expected to grow even more complex and heterogeneous. Optimizing performance and achieving portability represent critical pathways to achieving highest computational efficiency, reducing energy consumption, and ensuring that applications remain adaptable and seamlessly portable across rapidly evolving heterogeneous architectures. This presentation will discuss the challenges and pathways for optimizing application performance and achieving portability in modern heterogeneous HPC systems. Drawing from recent experiences in optimizing legacy applications, developing new simulation frameworks, and integrating data analysis pipelines, we will highlight strategies for leveraging multiple levels of parallelism—both within and across nodes—while maintaining a delicate balance between performance and portability. Key topics include scheduling libraries and autotuning, scalable domain decomposition, and runtime scheduling of workflows that integrate AI, data management, and simulations. The discussion will conclude with pathways for the effective and productive exploitation of parallelism and heterogeneity in next-generation HPC systems.
Bio
Florina Ciorba is Associate Professor at the University of Basel, Switzerland, where she heads the High Performance Computing lab she established in 2015. She received her PhD degree on algorithms design for parallelizing loops from the National Technical University of Athens, Greece. She and her team specialize in methods, tools, and techniques for performance, resilience, reproducibility, security, sustainability, and autonomous operation of HPC and AI/ML applications.
She is a member of various professional organizations (ACM Senior & Life Member, IEEE and IEEE Computer Society member), fora and boards (OpenMP ARB, MPI Forum, Energy Efficiency HPC WG, HiPEAC, SPEC HPG). She is a Founding Board Member and PI of UniBas node for the SKACH project (funded by SERI [https://skach.org/team/](https://skach.org/team/)), part of the Swiss Consortium of the SKAO. She is also a founding member of the IDEAS4HPC Association – a Swiss chapter of WHPC ([www.elle-it.ch](http://www.elle-it.ch/)). She has held various leadership positions in Parallel Computing and HPC conferences and workshops over the last decade.
Prof. Ciorba published almost a hundred peer-reviewed publications on high performance computing topics, and won best paper awards at several conferences. Her most recent contributions are in the area of autotuning with scheduling and load balancing libraries, energy-efficient cosmological simulations at extreme scales, and autonomy loops for monitoring and operational data analysis. More information at hpc.dmi.unibas.ch.
multicore.world 