ISC 2025 – WHPC Tech Talks

WHPC Poster Chair, Finalist – ISC25 Best Research Poster

Ayesha Afzal is a researcher at the Erlangen National High Performance Computing Center (NHR@FAU), with a background in computational and electrical engineering. Her PhD research focuses on analytic performance models, performance tools, and parallel simulation frameworks. She is involved in HPC initiatives like KONWHIR and NHR’s energy efficiency projects. Ayesha holds leadership roles in the IEEE Computer Society as vice chair of the Germany Section Chapter and Region 8 Area 2, and she founded the NHR Women in HPC chapter. She organizes key workshops such as EESP at ISC and PERMAVOST at ACM HPDC, and actively contributes to the HPC community as a reviewer, speaker, and program committee member. Her accolades include the ISC PhD Forum Award 2021, IEEE TPDS Best Paper Runner-up 2023, SC PMBS Best Short Paper 2023, and ISC Research Poster Award 2025. She was named to the Top 100 Future Leaders List (2022–2024) and won WeAreTheCity’s Global Award in 2023.

WHPC ISC25 – Partner Manager (Incoming) + WHPC Treasurer, and Chapters and Affiliates

Eleanor Broadway is an Applications Consultant at EPCC, the University of Edinburgh. She works in the team supporting the UK’s national supercomputing service, ARCHER2, providing in-depth support for users, implementing service improvements and developing training. She is also involved in a variety of technical research projects centred around optimising AI frameworks, porting applications to GPU and energy efficiency. Eleanor also regularly contributes to conferences and community events as an active volunteer for the Women in HPC (WHPC) community and to help promote HPC Outreach.

WHPC BoF Chair and Representing WHPC Central European Chapter

Karina Pešatová, MBA, leads the Training and Education Department at IT4Innovations National Supercomputing Centre in the Czech Republic. With an MBA in PR and Communication, she brings over eight years of experience in public relations, during which she managed and later led the Communication Department at IT4Innovations. Karina serves as the training champion for NCC Czechia in the EuroCC 2 project, contributes to the establishment of the pan-European EUMaster4HPC program, participates in the EuroHPC Virtual Training Academy establishment, and coordinates the Visegrad Fund-supported “Superheroes 4 Science” project, which aims to inspire and educate the next generation in HPC. Karina is one of the founding members of the Central European Chapter of Women in HPC and the current leader.

WHPC ISC25 Partner Manager (Outgoing) + WHPC Tech Talks Chair

Cristin Merritt, Chief Marketing Officer (CMO) at Alces Flight, boasts over 20 years of enterprise technology experience, with 7+ years dedicated to High Performance Computing (HPC). Specialising in navigating technology market trends, Cristin’s work explores the convergence of technical and social sustainability in the drive towards successful supercomputing solutions. Her tenure at Alces Flight has seen her spearhead intricate integration projects, three of which garnered global acclaim for their achievements in multi-platform (hybrid) HPC. Cristin’s leadership has cemented Alces Flight as a trailblazer in HPC integration and managed services, delivering solutions tailored to current and future workload demands.

An active contributor to the HPC community, Cristin leads as Chair for Business Development on the Women in HPC (WHPC) Executive Committee, while also managing WHPC’s social media presence. Additionally, she serves as an editor for the Supercomputing Conference (SC) social media channels and runs their annual Women’s History Month profile project.

Originally from Panama City, Florida, Cristin holds a degree in Classics from the University of Florida. She now proudly resides in Oxford, United Kingdom, with her husband, son, and a menagerie of beloved pets.

WHPC Travel Fellow

Sarah Johnston is a third-year astronomy PhD student in the Institute for Computational Cosmology at Durham University. She holds an MPhys in Astrophysics from University of St Andrews and is passionate about engaging with HPC and harnessing the full power of modern computing in astronomy.

Her PhD centres around implementing a GPU offload in the SWIFT cosmology code, where she is currently developing the GPU port for gravity calculations. Having already demonstrated a proof of concept, she is currently working on optimisation and effective implementation within SWIFT’s task-based system. She also works separately on constraining mixed dark matter cosmologies using large-scale simulations.

Beyond her research Sarah is actively involved in outreach and skill development. She leads research groups for young people to get hands-on with coding and simulations and facilitates training for other researchers, helping them develop the computational expertise needed for modern astrophysics.

Poster Title:

GPU Offloads for Gravity Calculations in SWIFT Cosmology Code

Abstract: 

To be compliant with modern heterogeneous HPC systems, large astronomy codes are needing to move towards GPU compatibility. This can be through total redesign or through partial offloading of key sections to GPU. SWIFT (SPH With Inter-dependent Fine-grained Tasking) is a versatile, open-source astronomy code used for a range of research areas in astronomy including galaxy formation, planetary impacts, and cosmology.  SWIFT utilizes task-based parallelism, which means that the workload is divided into independent tasks that can be executed concurrently, maximizing the use of available CPU resources, and is optimised for memory-intensive large CPU-only clusters. A significant portion of SWIFT‚Äôs runtime is dedicated to gravity calculations. In gravity n-body codes, each particle (representing a celestial object) interacts with every other particle based on gravitational forces, making the calculations computationally intensive. However, the repetitive and non-interdependent nature of these n-body interactions makes them ideal candidates for GPU acceleration.

In this work, we build on the existing SWIFT code by replacing specific CPU-based gravity calculation functions with new GPU kernels, minimizing disruption to the rest of the code while preserving the task-based parallelism. This creates a new hybrid C and CUDA version of the code which transfers gravity calculations to the GPU, freeing up the CPU to carry out the other tasks.

Our GPU-accelerated gravity kernels achieve high accuracy, with less than 1% deviation from CPU results below the Nyquist frequency. We also successfully produce nearly identical final particle distributions to the CPU-only implementation. Furthermore, the utilisation of GPUs allows for a redistribution of the gravity calculations meaning more interactions can be carried out using direct particle-particle summations which are more accurate, leaving only inexpensive multipole approximations to calculate on the CPU.

WHPC Travel Fellow

I am a researcher at IT4Innovations, specializing in deep learning and high-performance computing (HPC) for scientific imaging, including medical imaging, scientific visualization, anomaly detection, and materials science. Leveraging GPU-rich HPC systems like the Karolina supercomputer, I accelerate AI model training and large-scale data processing using optimized containerized workflows (Apptainer, Conda) and multi-GPU parallelization. At ISC25, I will present my work on SEM image segmentation for aluminide diffusion coatings, employing synthetic data generation with Blender and advanced models like U-Net and Swin UNETR, achieving high accuracy and training speedups via PyTorch and Slurm. My streamlined workflows span DICOM preprocessing to VR visualization. With expertise in HPC systems and PyTorch, I drive scalable machine learning solutions. As a WHPC Travel Fellow, I aim to showcase HPC innovation, mentor women in STEM, promote diversity in tech, and contribute to the HPC community through collaboration and knowledge-sharing.

Poster Title:

Deep Learning Approaches for Scanning Electron Microscope Image Analysis of Slurry Coatings

Abstract:

This study focuses on using deep learning to analyze Scanning Electron Microscope (SEM) images of aluminide diffusion coatings deposited on steel via the slurry route. The goal is to determine the thicknesses of the Fe2Al5 and FeAl coating layers, as well as the characteristics of pores and chromium precipitates. Due to challenges like imaging artefacts, noise, and overlapping features, a deep learning-based SEM image segmentation model using U-Net architecture was developed. Ground truth data were generated with the Weka segmentation plugin in ImageJ, refined manually, and augmented with synthetic data from Blender 3D. The model, trained on both synthetic and real SEM data, achieved mean dice scores of 98.7% ± 0.2 for Fe2Al5, 82.6% ± 8.1 for pores, and 81.48% ± 3.6 for precipitates. The method was applied to evaluate SEM images of coatings from three slurry compositions, revealing that slurries without rheology modifiers resulted in thicker Fe2Al5 layers. Coating thickness did not affect the relationship between outward and inward diffusion Fe2Al5 layers, with thinner coatings showing lower pore and chromium precipitate fractions. A high speed up is achieved by rendering synthetic images on HPC Karolina cluster. Also, a speed of 6.26x is achieved by performing distributed training on 8 GPUs instead of single GPU.

WHPC Distinguished Speaker

Buket joined Irish Centre for High-End Computing (ICHEC) in 2012 as a Computational Scientist. At ICHEC, she has contributed to user support activities as part of the National Service and worked on various PRACE and EuroHPC projects within Performance Engineering. She has also participated in graduate-level teaching as part of the Training and Education programme. Before joining ICHEC, she received Ph.D. from the Hamilton Institute, Maynooth University and M.Sc. in Computational Science and Engineering, Istanbul Technical University.

Buket is a member of the Women in HPC Ireland Chapter committee. Her primary interests include performance engineering particularly in optimising complex hybrid HPC workflows.

Tech Talk Abstract:

As HPC transitions from traditional applications to hybrid workflows, performance engineering is evolving to address new challenges. Furthermore, integration of traditional HPC, AI and Quantum Computing shifts the focus of performance analysis to optimising complex workloads and understanding unique needs of hybrid quantum HPC systems. My talk will cover these emerging shifts and expectations from performance engineering.

I’ve personally observed ongoing changes in HPC throughout my experience at ICHEC. I’ve worked on diverse projects, identifying bottlenecks in traditional HPC applications within PRACE [1] and EuroCC Academic Flagship [2] programme and later contributed to EuroHPC IO-SEA project [3], which focused on developing novel data management and storage platform for Exascale Computing. Currently, EuroCC Digital Innovation Flagship programme focuses on implementing complex domain-specific workflows, including Quantum Computing [4].

Meanwhile, HPC systems are growing larger and handling more complex datasets. The applications are becoming more complex, with workflows that need to process data in multiple stages. To ensure these systems work as needed, performance engineers need to have a broader set of skills such as data management, workflow automation, pipeline development and understanding emerging technologies.

As a result, performance engineers are required to integrate tools and techniques from the entire HPC ecosystem to deliver optimised solutions. To achieve that, it’s essential to be exposed to different disciplines, dedicate to continuous learning, collaborate with skilled colleagues, gain practical experience and work with a workforce that has a diverse skill set.

[1]. https://prace-ri.eu/
[2]. https://www.eurocc-ireland.ie/
[3]. https://iosea-project.eu/
[4]. https://www.eurocc-ireland.ie/dif

WHPC Distinguished Speaker

Jay Lofstead is a Principal Member of Technical Staff at Sandia National Laboratories. His research interests focus around large scale data management and trusting scientific computing. In particular, he works on storage, IO, metadata, workflows, reproducibility, software engineering, machine learning, and operating system-level support for any of these topics. Broadly across these topics, he is also deeply interested in ethics related to these topics and computing in general and how to drive inclusivity across the computation-related science domains. Dr. Lofstead received his Ph.D. in Computer Science from the Georgia Institute of Technology in 2010.

Representing JuWinHPC

Anna Lührs has a background in applied mathematics and computer science. She started her career at the Jülich Supercomputing Centre developing segmentation software for ultra-high resolution brain images on a GPU cluster. She soon shifted her focus towards project management, research coordination and science communication, initially for the Human Brain Project, one the largest EU-funded projects with more than 100 project partners and a total duration of 10 years. After the successful completion of this project, she joined the new Office for (Inter)national Coordination and Networking at the JSC, for which she also acts as deputy lead. She is co-founder and member of the organisation team of “Jülich Women in High Performance Computing” (JuWinHPC), the local chapter of the international organisation “Women in HPC”.

Representing WHPC MAR Chapter

Marta Garcia-Gasulla is a group leader at the Barcelona Supercomputing Center (BSC) in the Computer Science Department, where she has worked since 2006. She leads the Best Practices for Performance and Programmability (BePPP) group, focused on bridging the gap between domain scientists and computer scientists through performance analysis, co-design insights, and the promotion of best practices. She earned her PhD in Computer Architecture from the Universitat Politècnica de Catalunya (UPC) in 2017, where she was also an associate professor from 2008 to 2013. Her research interests include load balancing, application performance, hybrid parallel programming, and programming models. Marta has contributed to several European projects including ExaFOAM, POP3, EUPilot, CEEC, and DEEP-SEA, and has collaborated with industry leaders such as NVIDIA, Intel, and IBM. She is also a strong advocate for diversity in HPC and a founding member of the MAR Women in HPC chapter.

Representing WHPC Ireland

I head up engagement at Northern Ireland HPC (NI-HPC), a Tier-2 National HPC facility jointly managed by Queen’s University Belfast (QUB) and Ulster University. Along with my colleagues from the Irish Centre for High End Computing (ICHEC) and Technological University Dublin (TUD), we have recently launched a Women in HPC Ireland chapter.

With a background in FMCG, Pharma, Education, and Government, I have always considered myself a ‘non-techie person’ but I have been captivated by the impact that HPC has on research and, therefore, the changing landscape of our future.

My role is about building community, connecting stakeholders, and translating the technology into real-life stories.

I am focused on getting to know our HPC users and their research and collaborating with our stakeholders to build connections and promote our facility, I am particularly interested in growing our user base in non-traditional disciplines and making HPC more accessible to everyone.

Representing WHPC+ Australia and New Zealand

Mark is a dedicated leader at Pawsey Supercomputing Research Centre, who prioritizes promoting gender equity in the workplace. He recognizes the benefits of diversity and inclusivity, which has led to increased representation of the communities they serve. Mark takes visible, accountable actions to support equity at Pawsey, believing that it leads to better business outcomes and employee satisfaction. He leads a team of 50 experts and supports over 2,000 researchers using Pawsey’s advanced technology. Mark actively promotes an inclusive workplace by offering mentorship, resources for training, and driving change.

Representing IDEAS4HPC

Marie-Christine has been in HPC and large-scale computing over the last 2 decades: 2010-2020: director of Intel Exascale Lab in Europe, partnering with CEA, CERN and BSC. 2003 – 2008, managing director of CSCS, the Swiss national supercomputing centre. 2008-2010, CMS Computing at CERN.

She acted as a member of the Scientific Committee of the IFPEN- Institut Français du Pétrole Energies Nouvelles- between 2010 and 2014. She received funding for 4 Horizon projects. Active in the Women in HPC worldwide organisation, she is chairing the Swiss chapter of WPHC for which she obtained funding from the Simons Foundation. Presently HPC and Exascale Adviser of the ICES Foundation in Geneva, she was elected member of the SATW in 2024.

Today, she is active as an independent consultant, in conference technology selection committees & white-paper editor & speaker. She holds a PhD in Plasma physics from EPFL and Exec MBA from IMD

Tech Talk Presenter, WHPC Travel Fellow 2023

Aleksandra is a Cybersecurity Analyst specializing in digital identity, data protection, and vulnerability management. Her technical journey expanded into high-performance computing (HPC) during her second Master of Science in Computational Engineering, where she explored the intersection of natural language processing (NLP) and generative AI in the science of science context.

In addition to her industry work, Aleksandra teaches advanced programming and data analysis as an Academic Lecturer. Passionate about mentorship, she actively volunteers as a coach and trainer, supporting the professional growth of individuals across diverse communities.

Aleksandra has presented at numerous international conferences and academic events, sharing insights on cybersecurity, HPC, data science, digitalization, and career development. Her work has been recognized through international certifications and honors for interdisciplinary impact.

With a strong commitment to innovation, Aleksandra strives to bridge cybersecurity, data science, and advanced computing to help shape secure, efficient, and resilient digital systems for the future.

Tech Talk Abstract:

In both business and scientific software, maintaining cybersecurity principles is becoming increasingly important. The consequences of neglect in this matter are becoming severe. A few examples of compromised HPC systems will be mentioned to prove this statement. Therefore, more and more effort is being devoted to protecting against cyber threats, although still not enough. The presentation will discuss the key challenges of protecting data, software, and HPC infrastructure. The speaker will present potential threats and ways to prevent them, taking into account the latest technologies that can pose both a threat and a promising way of defense (e.g., AI algorithms). The audience will also learn about well-known solutions and tools that can be implemented in HPC, such as IAM, monitoring, network security, or job management. The “secure by design” principle will also be mentioned in the context of designing computing clusters from scratch and several other security concepts. Let’s connect the dots at ISC25, including cybersecurity as an emerging trend!

Tech Talk Presenter

Antonia Maar is a Research Analyst at Intersect360 Research, based in Germany. She focuses on high performance computing (HPC) and artificial intelligence (AI), delivering data-driven insights that support strategic decision-making across the industry. Antonia brings cross-sector experience from roles in both consulting and industry. She holds a Master’s degree in Management from the Frankfurt School of Finance & Management and completed an International Honors Program at Stanford University, where she focused on HPC, AI, and Symbolic Systems.

At Intersect360 Research, she contributes to industry analyses that inform the decisions of vendors, policymakers, and end users navigating the fast-evolving HPC and AI landscape. Antonia is particularly interested in how AI is both transforming and leveraging the HPC ecosystem—from system architecture and investment priorities to adoption trends and policy developments—and is passionate about making complex technologies accessible to a broad range of stakeholders.

Tech Talk Abstract:

One year ago, I never would have expected to find myself working in high performance computing. With a background in business and no formal technical training, HPC was a field I admired from the outside — impressive, but not something I thought I could be part of.  That changed after taking a course at Stanford and meeting someone who saw potential beyond the typical checkboxes. That moment opened a door I didn’t know existed.

This talk shares my transition from business studies into the HPC and AI space, and why stories like this matter. As the industry faces a growing talent gap, we must rethink how we attract and support new talent. This starts with people who are willing to give chances, and with showing that there’s more than one way to enter this field.

I’ll reflect on the importance of sponsorship, visible role models, and the need to broaden our understanding of who belongs in HPC.  The field’s future depends not only on engineers, but also on those with unconventional paths and non-technical backgrounds. Roles in analysis, strategy, and communication require just as much immersion — and they contribute in ways that are equally essential. Because impact in HPC doesn’t come from code alone — it also comes from curiosity, fresh perspectives, and courage to take an unexpected path.