Galvez J1, Oliveira M2, Sinha M2, Yang E2
1Nci, , Australia, 2ACCESS-NRI, Acton, Australia
Biography:
Jorge is an HPC specialist who has experience in the computational chemistry and geophysical fluid dynamics fields. His main strength is GPU programming and porting codebases from CPU to GPUs.
Micael leads the Software Transformation Team – bringing the specialist software engineering skills required for the ACCESS modelling framework to take advantage of advances in computing hardware, and to position ACCESS for future changes in technology. Micael enjoys designing and developing complex pieces of software and enabling new science through software engineering best-practices and use of state-of-the-art hardware architectures.
Micael has a PhD in Computational Physics, from the University of Coimbra, Portugal, 2009. Micael is interested in all aspects of software development and high-performance computing applied to fundamental scientific research. He has a lot of experience in developing massively parallel scientific codes, and is a strong advocate for well written, modular, reusable, easy to maintain software.
Outside of work, Micael enjoys reading, running and playing board games.
Edward is a Senior Research Software Engineer of the Software Transformation team specialising in Machine Learning. He has five years’ experience in high-performance computing and scientific software development, specialising in accelerating computational workflows, leveraging GPU and MPI parallelisation, and supporting Machine-Learning pipelines. Edward’s experience spans administering HPC systems and enabling research through automation and optimization of computationally intensive tasks.
Edward has a Master of Engineering Science (Monash University, 2021) with a thesis on Development of a scalable parallel SPH code for large-scale geotechnical applications.
Edward’s current role involves implementing and maintaining Machine-Learning capabilities in ACCESS’ climate forecasting codes, as well as supporting software improvements like GPU acceleration and improving software maintainability. He enjoys the ample opportunities to problem solve and be creative; learning about the latest HPC hardware and software technologies; and working on the cutting-edge of Machine Learning.
Outside of work, Edward likes to go cycling (either road or gravel), hiking, and playing the piano. Lately, he also has also been getting into bird photography!
Dr Manodeep Sinha is a Principal Research Software Engineer in the Software Transformation team. He is well-known within the astronomy community for his “wizardry” in software optimisation, and finds himself working closely with the software creators, which he says is due to his “stubbornness” and refusal to yield to the computer!
Manodeep is a champion for systemically excluded people and a strong advocate for improving practices using an intersectionality framework. These principles manifest as his commitment towards open-source software and “Open Science”, improving gender equality practices, and increasing the recognition of the term “RSE” and the critical contributions of RSEs towards modern research.
Manodeep has a Bachelor in Electrical Engg. and has gained a PhD in Astronomy & Astrophysics from The Pennsylvania State University in 2008 Thesis Title: Hot Halo Gas in Numerical Simulations of Galaxy Mergers.
Abstract:
At almost 70 years old, Fortran remains central to HPC, being one of the most used programming languages for computationally intensive applications. Fortran's longevity reflects its continuous evolution to support modern programming paradigms, such as array, object-oriented, and parallel programming (shared and distributed memory). More recently, standardized C interoperability has made it simple to write Fortran programs that can be interfaced with other languages such as Python and Julia. Yet, most modern features of Fortran remain underappreciated outside the Fortran community. In this BoF, we will explore why Fortran is a good choice for new development in 2026. Topics include the evolving ecosystem (standard library, fpm), GPU acceleration via standard parallelism, agentic AI tooling for Fortran, and teaching Fortran to developers from other languages.
This session is intended for HPC practitioners, computational scientists, compiler and toolchain developers, and educators, both experienced Fortran users and developers from other backgrounds evaluating Fortran for new projects.
The session will start with four 5-minute presentations covering ecosystem, compiler and GPU support, and education (20 min), followed by a moderated roundtable with active solicitation of floor contributions (30 min). Finally, the session will conclude with a wrap-up synthesising common themes, open questions, and next steps (10 min).
By the end of the session, attendees will gain a current picture of Fortran and its ecosystem, practical insights into GPU integration, awareness of best practices, and new connections across the global Fortran community. The session will identify concrete collaboration opportunities on educational materials, scientific libraries, and tooling.
