The International Conference for High Performance Computing, Networking, Storage, and Analysis
The International Conference for High Performance Computing, Networking, Storage, and Analysis
Authors: Charles Doutriaux (Lawrence Livermore National Laboratory (LLNL)), Christine Sweeney (Los Alamos National Laboratory (LANL)), David Lugato (Commissariat a l'Energie Atomique), Rusty Davis (Los Alamos National Laboratory (LANL)), Ron Oldfield (Sandia National Laboratories)
Abstract: As AI and ML technologies converge with computational science and engineering, practitioners face emerging challenges in developing and deploying effective workflows. This BoF session invites SC25 attendees to discuss critical issues including workflow composition and orchestration, containerization, robust data management, and AI integration with simulation workflows and infrastructure. We will discuss a variety of AI workflows that include LLMs, science models, and AI agents. Through a short keynote, lightning talks, small group discussion, and short audience polls, this BoF is designed to be interactive and focus on audience interests. Join us to share experiences and identify key challenges.
Long Description: As artificial intelligence (AI) and machine learning (ML) technologies increasingly intersect with computational science and engineering, practitioners are confronted with a range of emerging challenges that impact the development and deployment of effective workflows. This birds-of-a-feather session invites researchers, developers, and industry professionals to engage in an open dialogue about the critical issues arising from this convergence. Key topics for discussion will include the necessity of robust data management and governance strategies, the challenges of ensuring data quality and integrity, model interpretability and transparency, and the integration of AI systems into existing computational infrastructures. We will place particular emphasis on the development and deployment of workflows that leverage High-Performance Computing (HPC) as a service to facilitate complex computations and large-scale data processing. Additionally, we will explore the role of containerization and workflow orchestration in optimizing AI/ML operations within computational frameworks. Participants will have the opportunity to share their experiences, explore innovative solutions, and collaborate on best practices for effective workflow development and deployment. By fostering a community of shared knowledge and support, this session aims to empower attendees to enhance their AI/ML workflows in the context of computational science and engineering. Join us to contribute to this vital conversation and help shape the future of AI and ML practices, emphasizing the strategic integration of HPC resources and the adoption of containerization and orchestration techniques as foundational elements of successful workflow implementation. The schedule will be divided between lightning talks, presentations and panel discussions.
Website: https://sites.google.com/view/sc25-bof-ai-workflows/home