LLM4HPC 2025

The 1st International Workshop on Foundational large Language Models Advances for HPC

to be held in conjunction with
ISC-HPC 2025

13 June, 2025
Hamburg, Germany

Introduction

Since their development and release, modern Large Language Models (LLMs), such as the Generative Pre-trained Transformer (GPT) model and the Large Language Model Meta AI (LLaMA), have come to signify a revolution in human-computer interaction spurred on by their high-quality results. LLMs have repaved this landscape thanks to unprecedented investments and enormous training models (hundreds of billions of parameters). The availability of LLMs has led to increasing interest in how they could be applied to a large variety of applications. The HPC community made recent research efforts to evaluate current LLM capabilities for some HPC tasks, including code generation, auto parallelization, performance portability, correctness, among others. All these studies concluded that state-of-the-art LLM capabilities have proven so far insufficient for these targets. Hence, it is necessary to explore novel techniques to further empower LLMs to enrich the HPC mission and its impact.

Call For Papers

Objectives, scope and topics of the workshop

This workshop objectives are focused on LLMs advances for any HPC major priority and challenge with the aims to define and discuss the fundamentals of LLMs for HPC-specific tasks, including but not limited to hardware design, compilation, parallel programming models and runtimes, application development, enabling LLM technologies to have more autonomous decision-making about the efficient use of HPC. This workshop aims to provide a forum to discuss new and emerging solutions to address these important challenges towards an AI-assisted HPC era. Papers are being sought on many aspects of LLM for HPC targets including (but not limited to):

• LLMs for Programming Environments and Runtime Systems
• LLMs for HPC and Scientific Applications
• LLMs for Hardware design (including non-von Neumann Architectures)
• Reliability/Benchmarking/Measurements for LLMs

Program

9,00AM-9,15AM: Opening Pedro Valero-Lara
9,15AM-10,00AM: Keynote: LLM-enabled swarm intelligent agents for resilient HPC infrastructures, Prasanna Balaprakash
10,00AM-10,30AM: First talk: Analysis of MPI Parallel Code Generated by GPT-4o, Rin Tanaka
10,30AM-11,00AM: Second talk: LLM & HPC:Benchmarking DeepSeek's Performance in High-Performance Computing Tasks, Patrick Diehl
11,00AM-11,30AM: Break
11,30AM-12,00PM: Third talk: Leveraging AI for productive and trustworthy HPC software: challenges and research directions, Pedro Valero-Lara
12,00PM- 1,00PM: Panel: LLM4HPC --Challenges and Opportunities
Moderator: Daniel Lee Nichols
Panelists: Jeffrey S. Vetter, Prasanna Balaprakash, Rin Tanaka, Patrick Diehl, Pedro Valero-Lara

Important Dates

Paper submission deadline : March 14 (FIRM DEADLINE), 2025
Notification of acceptance : March 31, 2025
Camera-ready papers due : May 16, 2025
Workshop day: June 13, 2025

Steering Committee

Jeffrey S. Vetter, Oak Ridge National Laboratory, USA

Rosa M. Badia, Barcelona Supercomputing Center, Spain

Franz Franchetti, Carnegie Mellon University, USA

Enrique Quintana Orti, Universitat Politecnica de Valencia, Spain

Abhinav Bhatele, University of Maryland, USA

Organizers (Contact us)

Pedro Valero-Lara (chair)
Oak Ridge National Laboratory, USA
valerolarap@ornl.gov

Harshitha Menon (co-chair)
Lawrence Livermore National Laboratory, USA
harshitha@llnl.gov

Konstantinos Parasyris (co-chair)
Lawrence Livermore National Laboratory, USA
parasyris1@llnl.gov

Daniel Lee Nichols (co-chair)
University of Maryland, USA
dnicho@umd.edu

Programme Committee

• Samuel Williams, Lawrence Berkeley National Laboratory, USA
• Hiroyuki Takizawa, Tohoku University, Japan
• Gokcen Kestor, Barcelona Supercomputing Center, Spain
• Prasanna Balaprakash, Oak Ridge National Laboratory, USA
• Rabab Alomairy, Massachusetts Institute of Technology, USA
• Johannes Blaschke, Lawrence Berkeley National Laboratory, USA
• Ramakrishnan (Ramki) Kannan, Oak Ridge National Laboratory, USA
• Olivier Aumage, INRIA, France
• Ignacio Laguna, Lawrence Livermore National Laboratory, USA
• Johannes Doerfert, Lawrence Livermore National Laboratory, USA
• Monil Mohammad Alaul Haque, Oak Ridge National Laboratory, USA
• Simon Garcia De Gonzalo, Sandia National Laboratory, USA
• Diego Andrade Canosa, University of A Coruna, Spain
• Tze-Meng Low, Carnegie Mellon University, USA
• Dario Garcia Casulla, Barcelona Supercomputing Center, Spain
• Michel Schanen, Argonne National Laboratory, USA
• Keita Teranishi, Oak Ridge National Laboratory, USA
• William F. Godoy, Oak Ridge National Laboratory, USA
• Damian Rouson, Lawrence Berkeley National Laboratory, USA
• Jens Domke, RIKEN Center for Computational Science(R-CSS), Japan
• Narasinga Rao Minskar, Oak Ridge National Laboratory, USA
• Sunita Chandrasekaran, University of Delaware, USA
• Arjun Guha, Northeastern University, USA

Manuscript submission

We invite submissions of original, unpublished research and experiential papers. Papers should be between 6 to 12 pages in length (including a bibliography and appendices, with two possible extra pages after the review to address the reviewer’s comments), formatted according to Springer’s Lecture Notes in Computer Science (LNCS). All paper submissions will be managed electronically via EasyChair.

Proceedings

All accepted papers will be published in the ISC-HPC Workshops 2025 proceedings by SpringerLink.

Best Paper Award

The Best Paper Award will be selected on the basis of explicit recommendations of the reviewers and their scoring towards the paper’s originality and quality.

Invited Speaker (Prasanna Balaprakash, Oak Ridge National Laboratory):

TBA

LLM-enabled swarm intelligent agents for resilient HPC infrastructures

Existing adaptive management and resource partitioning strategies for resilient HPC infrastructures typically rely on static, expert-developed rules and centralized control mechanisms. While mainstream artificial intelligence (AI) methods promise dynamic, real-time resource management, their scalability and effectiveness remain limited when confronting the unique complexities and scales inherent to resilient infrastructures. In this talk, Prasanna will present the U.S. Department of Energy funded SWARM project that seeks to address this gap by investigating distributed intelligence, specifically harnessing the Large Language Model (LLM)-enabled swarm intelligence (SI). This innovative approach aims to provide robust, high-performance, resilient, and fault-tolerant management of scientific workflows. These workflows extend across diverse resources, from edge devices situated near sensors and instruments, through extensive wide-area networks, to leadership-class computational systems. The overarching objective of SWARM is to develop LLM-enabled SI agents for resilient infrastructure capable of rapidly recovering from failures, dynamically adapting to environmental changes, optimizing resource utilization, and minimizing workflow execution times for scientific communities.

Prasanna Balaprakash is the Director of AI Programs and a Distinguished R&D Scientist at Oak Ridge National Laboratory. Balaprakash’s research interests span AI, machine learning, optimization, and high-performance computing. He received the U.S. Department of Energy's Early Career Award in 2018. He co-leads the research and development of DeepHyper, a scalable automated machine learning package for developing trustworthy and energy-efficient AI models on DOE leadership-class systems. He was an R&D Group Leader at Argonne. Balaprakash received his PhD in 2010 from CoDE-IRIDIA (AI Lab), Université Libre de Bruxelles, Brussels, Belgium, where he was honored with Marie Skłodowska-Curie and F.R.S-FNRS Aspirant fellowships from the European Commission and the Belgian-French Community's National Fund for Scientific Research, respectively.

Registration

Information about registration at ISC-HPC 2025 website.