AGENT4SC focuses on the systems foundations required to operationalize agent decisions that directly affect allocations, system health, data movement, energy consumption, and scientific outcomes. This demands integrated data and metadata infrastructures spanning scalable databases, vector stores, caching layers, provenance services, and knowledge graphs, all designed to remain performant and auditable under high concurrency. At scale, even minor errors, hallucinations, or blind spots can escalate into megawatt-hour waste, irreproducibility, and compromised scientific validity. Therefore, this workshop provides a forum to advance scalable architectures, cross-continuum coordination, evaluation frameworks, verification and mitigation strategies, provenance and observability mechanisms, performance analysis, and lessons from real deployments. The workshop builds on active efforts in agentic AI for science across national laboratories, academia, and industry, reflecting an emerging and rapidly growing community.
Graph-based technologies, including knowledge graphs, ontologies, graph databases, graph analytics, and graph machine learning, are emerging as powerful unifying abstractions across all stages of the scientific research lifecycle. This workshop explores how these technologies can be applied to represent and integrate knowledge, model workflow dependencies, enable semantic interoperability across heterogeneous datasets, and support resource discovery, reasoning, and orchestration in distributed computing environments, including HPC, Cloud, Edge, and IoT infrastructures spanning the computing continuum. Beyond infrastructure and middleware concerns, the workshop welcomes contributions from any domain of science that employs graph-based methods as a primary tool for data integration, analysis, or knowledge discovery.
The First Workshop on Research Software Supply Chain Security (RS3) brings together research software engineers, scientists, infrastructure operators, and cybersecurity researchers to tackle a neglected but increasingly important problem: the security of the research software supply chain. Modern science runs on software, and that software is assembled from many components whose trustworthiness is rarely easy to judge. RS3 will give participants a place to compare experiences and discuss where current practice falls short. The workshop will examine how stronger security can fit the realities of scientific work rather than fight against them. Attendees will come away with a clearer understanding of the risks in research software ecosystems, a vision for next steps, and new connections with others working on the same problems.
Scientific automation is at the forefront of enabling faster, more reproducible discoveries in AI, exascale computing, and distributed environments. The WISDOM workshop provides a platform for researchers, developers, and practitioners to discuss advances in autonomous workflow design, intelligent data management, and scalable and agentic automation techniques. We are especially interested in innovative methods that integrate AI/ML to enhance the flexibility, efficiency, and reliability of scientific processes across heterogeneous environments.
The Global Research Platform (GRP) is an international scientific collaboration that aims to create one-of-a-kind advanced ubiquitous services that integrate resources around the globe at speeds of gigabits and terabits per second. GRP focuses on design, implementation, and operation strategies for next-generation distributed services and infrastructure to facilitate high-performance data gathering, analytics, transport, computing, and storage, at 100 Gbps or higher. GRP actively works with partners in North America, Asia, Europe, and South America to customize international fabrics and distributed cyberinfrastructure to support data-intensive scientific workflows.
AI and data capability programs are increasingly central to eScience, but many fail to create durable change because they focus on tools rather than the interaction of technological, human, and organisational conditions that make adoption stick. This workshop brings together practitioners, experts, and researchers from academia, government, industry, and faith-based organisations to share real-world case studies of AI capability uplift, with attention to what works, what fails, and why. Through invited talks, peer-reviewed extended abstracts, posters, and an interactive closing session, participants will examine governance, trust, human–AI collaboration, and the cultural and technological narratives that shape transformation. The workshop will also produce a shared AI Capability Uplift Playbook to capture transferable lessons for the broader eScience community.
The workshop focuses on the integration of data-driven methodologies, high-performance computing (HPC), and physics-based models for atmospheric science applications, with particular emphasis on air quality, climate interactions, and anthropogenic impacts. In recent years, atmospheric modelling has undergone a paradigm shift driven by the availability of large datasets (satellite observations, sensor networks), advances in machine learning, and the development of scalable computing infrastructures. These developments have enabled the emergence of hybrid approaches that combine numerical modelling, statistical inference, and artificial intelligence, leading to improved predictive capabilities and new scientific insights. The workshop aims to explore this transition from traditional modelling frameworks to integrated eScience approaches, emphasizing reproducibility, scalability, and interoperability.
The INSTIL workshop focuses on ICT-enabled citizen science and its role in supporting participatory and data-driven scientific research. The workshop brings together researchers and practitioners working on digital platforms, sensor networks, and collaborative tools that enable citizens to actively contribute to scientific processes. A particular emphasis is placed on applications related to environmental monitoring, climate science, and climate change, where citizen-generated data and participatory sensing can complement traditional scientific infrastructures. In this context, emerging approaches such as citizen observatories are gaining increasing relevance. INSTIL aims to foster interdisciplinary dialogue across computer science, environmental sciences, and social sciences, highlighting how digital technologies can support citizen engagement and contribute to addressing complex societal challenges.
The development of computational models for simulating the behaviour and evolution of natural and engineered systems continues to play a central role in advancing science and engineering, and is currently being reshaped by the convergence of High Performance Computing (HPC), data-intensive methodologies, and artificial intelligence. The advent of heterogeneous and exascale architectures, coupled with rapid progress in GPU computing and distributed workflows, is transforming the way simulations are designed, executed, and integrated with large-scale data. In this context, traditional numerical approaches such as Finite Element, Finite Difference, and Particle-In-Cell methods are increasingly complemented by data-driven and hybrid techniques, including machine learning, physics-informed neural networks, surrogate models, and bio-inspired algorithms, enabling new capabilities for addressing complex multi-scale and multi-physics problems. These advances are also fostering the development of digital twins and real-time simulation frameworks, while raising new challenges in terms of scalability, portability, performance, and energy efficiency. The HPCMS Workshop at eScience 2026 aims to bring together a multidisciplinary community of researchers and practitioners to discuss recent advances and emerging trends in modelling and simulation, with particular emphasis on the integration of HPC and AI, scalable and heterogeneous computing, data-driven workflows, and sustainable computing practices. The workshop will also explore software ecosystems and programming models, including MPI, OpenMP, SYCL, and CUDA, as well as performance modelling, optimization techniques, and novel hardware approaches such as FPGAs, neuromorphic computing, and quantum computing. By fostering cross-disciplinary interaction across domains ranging from engineering and physical sciences to life sciences and socio-economic systems, the workshop seeks to promote innovative solutions to complex real-world challenges and to advance the next generation of computational science methodologies.
Emerging and future computational workloads are combining traditional HPC applications with tools and techniques from the scale-out data analytics and machine learning community. Getting these technologies to co-exist and interoperate to advance scientific discovery is a daunting task with few known good solutions. In general, constructing these workflows has the potential to create pitfalls and incompatibilities that limit adoption. Formalizing the steps necessary for an application or data processing pipeline is increasingly popular and necessary. Requirements for reproducibility artifacts for publishing venues are also driving this formalization. The processes and infrastructure to accomplish these requirements are frequently bespoke or custom for a particular research area. All of these formalization activities can be described as workflow systems. Existing off-the-shelf tools address a distributed environment fairly well, but are not complete solutions and do not address the scale-up community much, if at all. Complicating managing workflows are the tasks of managing data both during workflow execution and then afterward as well as offering authentication and data security for shared data sets. With some data, such as climate simulation output, being subject to intense scrutiny, it becomes crucial to offer open data that can be verified as authentic by means of encrypted creator identities, and accessible only to people with a need to know. Sharing and analyzing the data knowing it is authentic while protecting the privacy of the creators is essential for reliable open science while protecting the identity of the scientists performing the work. This workshop seeks to explore ideas and experiences on what kinds of infrastructure developments can improve upon the state of the art. Explorations of component packaging via containers and virtual machines, automation scripting, deployment, portability builds, and system support for these and other relevant activities are key infrastructure. Provenance collection, exploration, and tracking are key for a well-documented scientific output. Using existing systems to achieve these goals via experiences is important for developing best practices that span application domains. Data privacy techniques such as multi-party encryption and differential privacy are important as well. Issues with managing large data sets and workflow intermediate data, particularly those intended to manage publicly accessed data for use and reuse are encouraged. New techniques and technologies that address reproducibility requirements are also requested. We seek work on all of these, and related, topics as well as position and experience papers looking to drive the conversation for practitioners and researchers in these spaces. This workshop contributes by sharing experiences and exploring the various technological infrastructure needs to support effective, convenient workflow systems and application composition structures and approaches across a broad spectrum of HPC environments from clusters to supercomputers to cloud systems.
