NHR PerfLab Seminar: From Data Traces to System Insight—Redefining I/O Understanding in HPC (June 24, online)
Speaker: Prof. Dr. Sarah Neuwirth, Full Professor of Computer Science, Johannes Gutenberg University Mainz
Date and time: Tuesday, June 24, 2025, 2:00 p.m. CEST
Zoom: https://go-nhr.de/perflab-seminar
Abstract:
As high-performance computing systems evolve to accommodate increasingly heterogeneous and data-intensive workloads, I/O behavior is emerging as a critical determinant of system efficiency, reliability, and scalability. This talk explores how rethinking I/O as a source of insight, rather than simply a performance constraint, opens new pathways for analysis, optimization, and design. I will discuss the growing complexity of I/O patterns across modern workflows, the challenges of achieving observability across deeply layered I/O stacks, and the importance of trace-driven methodologies for ensuring reproducibility and explainability.
We will examine how these factors influence performance variability, system design trade-offs, and future benchmarking practices, particularly in the context of shifting architectural paradigms, emerging consistency models, and the increasing role of adaptive, multi-tiered storage. The goal of this talk is to offer a perspective on I/O that is both system-aware and forward-looking, highlighting its central role in shaping the next generation of sustainable and transparent HPC infrastructures.
Short Bio:
Sarah Neuwirth is a Full Professor of Computer Science and Chair of the “High Performance
Computing and its Applications” research group at Johannes Gutenberg University Mainz. 
She also serves as the co-director of the NHR South-West HPC Center. In 2018, Sarah completed her PhD in computer science at Heidelberg University. Her research interests include parallel I/O and storage systems, modular supercomputing, performance modeling and analysis, optimization, reproducible benchmarking, and parallel programming models. For her outstanding contributions to HPC, Sarah was awarded the “2023 PRACE Ada Lovelace Award for HPC” and the “ZONTA Science Award 2019.” She has participated in numerous research collaborations as co-PI, including working with Jülich Supercomputing Centre (DEEP Project Series, EUPEX), LLNL, BITS Pilani Goa Campus, ORNL, and Virginia Tech. Sarah has held various leadership positions in HPC conferences and workshops over the past years.
For a list of past and upcoming NHR PerfLab seminar events, please see: https://hpc.fau.de/research/nhr-perflab-seminar-series/