Workshops

DRAI: Data Readiness for AI

Data fuels Artificial Intelligence (AI) applications. Hence, collecting and maintaining high-quality data that has been evaluated to be ready for AI is critical to creating new large AI models that assist in enabling scientific breakthroughs across domains. However, there are several challenges in achieving AI readiness for research data. This workshop aims to enable discussions on novel and efficient methods for collecting and preparing data for AI, metrics to quantify data readiness, frameworks for improving data readiness,  assessments of the impact of data on AI model performance, and existing challenges and caveats in managing and transforming historical and new data from various science domains into AI-ready data.

EMS: Embedded Multicore Systems

ICPP-EMS focuses on embedded multi-core computing, a key area for both industry and academia. While these systems are crucial for future designs, challenges remain in applications, programming models, architectures, memory, and software tools. This forum brings together researchers to discuss advances in AI/embedded compilers, memory, DSP/GPU systems, and multi-core programming. Topics include CPU/GPU/DSP/DLA/DPU/PIM fusion, optimization with OpenCL, CUDA, Halide, SPIR-V, MLIR, TVM, LLVM, and more. Work-in-progress, case studies, and visionary ideas are welcome.

FTXS: Fault-Tolerance for HPC at Extreme Scale

Increases in the number, variety, and complexity of components required to compose next-generation extreme-scale systems mean that systems will experience significant increases in aggregate fault rates, fault diversity, and fault complexity. Additionally, the growing importance of AI/ML workloads, increasing system heterogeneity, and the emergence of novel computing paradigms (neuromorphic, quantum) introduce fault tolerance issues that the research community has just begun to address. Given the continued need for research on fault tolerance in extreme-scale systems, FTXS 2025 will present an opportunity to share, discuss and evaluate innovative ideas on fault tolerance.

Link to FTXS Workshop Page

GRAND: Graph Reasoning and AI for Network Discovery

The integration of high-performance computing (HPC) with dynamic and static graph analysis has revolutionized artificial intelligence (AI) by enabling scalable solutions for complex challenges in domains like social networks, and bioinformatics. This workshop will explore leveraging HPC for efficient analysis of large-scale, evolving graph datasets, focusing on topics like algorithm optimization, distributed processing for graph neural networks (GNNs), and dynamic graph updates. Attendees will gain insights into practical AI applications. This event fosters collaboration among researchers and industry leaders, accelerating innovation in graph analytics and AI.      

Harvest: Applications of HPC and AI in Agriculture 

Modern AI systems powered by high-performance computing have revolutionized how we interact with technology, and their combined impact on various science domains like agriculture has yet to be fully explored. Over the past decades, computer scientists, computational scientists, agronomists, and researchers in agriculture have dedicated their efforts towards empowering farmers by ensuring the latest technological advances are available to them. This workshop aims to bring together researchers and software/hardware designers from academia, industry, and national laboratories involved in designing HPC-powered AI-enabled systems for agriculture and how it can be leveraged to improve efficiency, accuracy, and accessibility for end-users.

WISDOM: Workflows, Intelligent Scientific Data and Optimization for automated Management

Scientific automation is key to accelerating discovery in AI, exascale computing, and distributed environments. The Workflows, Intelligent Scientific Data and Optimization for Automated Management (WISDOM) workshop will bring together researchers, developers, and practitioners to explore innovative methods that leverage workflows and data to automate scientific processes across the computing continuum. It will address challenges such as workflow design, data-driven automation, metadata strategies, and integrating machine learning for intelligent decision-making. WISDOM aims to advance scientific automation and foster autonomous discovery, encouraging contributions to automated workflows, self-driving labs, and autonomous research.

Link to WISDOM Workshop Page