Dr Slava Kitaeff1,2, Luc Betbeder-Matibet2, Jake Carrol3
1Monash University, Australia, 2The University of New South Whales, Sydney, Australia, 3The University of Queensland, Brisbane, Australia
Biography:
Dr. Slava Kitaeff is a nationally recognised leader in high-performance computing, data-intensive science, and large-scale digital research infrastructure. As Associate Director eResearch (HPC) at Monash University, he led major national initiatives including the National ML/AI Research Platform and the Australian Characterisation Commons at Scale Platform. As the Australian SKA Regional Centre Program Lead at UWA & CSIRO, he directed the Design Study Program and chaired the International SRC Architecture Working Group. Dr. Kitaeff was also Research Associate Professor for HPC and Data-intensive science at the University of Western Australia, where he developed novel methods for high-performance computing applications and data-intensive processing pipelines. He is an adjunct professor of HPC at the School of Engineering of Monash University. He has held senior leadership roles for over a decade across multiple institutions, focusing on research computing, digital innovation, and software engineering education.
Luc Betbeder-Matibet is a nationally recognised subject matter expert in High-Performance Research Computing, Research Data practices and shared Research Infrastructure services. He is Vice-President of AeRo (Australia's eResearch Organisation) and co-chair of the eResearch Australasian peak Annual Conference. He is an Adjunct Senior Lecturer in UNSW Faculty of Medicine Centre for Big Data and has been a Visiting Scientist with the Data61 Visual Analytics Team in Australia's National Science Agency CSIRO. He has held director-level roles for 15 years in higher-ed ICT and eResearch. Luc is currently the Executive Director Research Technology Services at UNSW.
Jake Carroll is the Director of The University of Queensland Research Computing Centre and is a Doctoral Candidate in the School of Electrical Engineering and Computer Science in the same institution. He has been involved in large scale architecture, systems and design of digital scientific research infrastructure for twenty years. He serves on numerous boards of advisory across various sectors encompassing corporate, health and defence related advanced computing domains. Carroll has a Bachelor of Information Technology (Honors) from Southern Cross University and an MBA from The University of Queensland.
Abstract:
The sustainability of research computing is emerging as a critical issue across institutions and national facilities alike. As AI workloads drive unprecedented demand for GPU infrastructure, and as energy prices, carbon goals, and physical constraints place new pressures on data centres, the research computing community must rethink how resources are provisioned, operated, and used.
This BoF invites participants from across the research computing landscape—operations teams, research leaders, infrastructure strategists, and power users—to discuss practical challenges and share emerging solutions. Key topics will include data centre sustainability strategies (e.g., commercial vs on-premise), power and cooling limitations, the energy cost of AI and large-scale workflows, and how user behaviour, software choices, and scheduling policies impact overall efficiency.
We will also consider what “sustainability” means beyond electricity: how funding models, training programs, usage culture, and policy levers can influence longer-term viability and equitable access. Can we track power usage like CPU hours? Should GPU jobs have sustainability guidelines? What new norms or collaborations are needed?
Whether your concern is carbon, cost, or capacity, this BoF aims to provide space for informed discussion, shared learning, and actionable next steps for making research computing more sustainable—without compromising performance or innovation.