EuroHPC SPACE CoE: Redesigning Scalable Parallel Astrophysical Codes for Exascale

2025-12-21

AI Summary

Exascale supercomputing architectures introduce severe performance bottlenecks in astrophysical simulations due to their extreme complexity and heterogeneity. Method: We refactor mainstream astrophysics codes—including GADGET and ENZO—introducing a portable heterogeneous programming model (based on SYCL/Alpaka) and a standardized scientific data protocol. Our approach unifies HPC, hardware vendor, astrophysics, and machine learning expertise to co-design a full-stack software ecosystem. Contribution/Results: The resulting toolchain delivers adaptive load balancing, distributed memory optimization, scientific ML–enabled analysis, high-performance visualization, and FAIR-compliant data management. Deployed at scale on LUMI and LEONARDO, it achieves efficient execution across hundreds of nodes and thousands of GPUs, delivering 3–8× performance gains. It enables real-time, PB-scale simulations with >10⁸ grid cells and multi-physics coupling, significantly advancing computational astrophysics at exascale.

Technology Category

• Machine Learning: Hardware-aware ML
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

High Performance Computing (HPC) based simulations are crucial in Astrophysics and Cosmology (A&C), helping scientists investigate and understand complex astrophysical phenomena. Taking advantage of exascale computing capabilities is essential for these efforts. However, the unprecedented architectural complexity of exascale systems impacts legacy codes. The SPACE Centre of Excellence (CoE) aims to re-engineer key astrophysical codes to tackle new computational challenges by adopting innovative programming paradigms and software (SW) solutions. SPACE brings together scientists, code developers, HPC experts, hardware (HW) manufacturers, and SW developers. This collaboration enhances exascale A&C applications, promoting the use of exascale and post-exascale computing capabilities. Additionally, SPACE addresses high-performance data analysis for the massive data outputs from exascale simulations and modern observations, using machine learning (ML) and visualisation tools. The project facilitates application deployment across platforms by focusing on code repositories and data sharing, integrating European astrophysical communities around exascale computing with standardised SW and data protocols.

Problem

• Redesigning legacy astrophysical codes for exascale HPC architectural challenges.
• Enhancing exascale simulations with innovative programming and software solutions.
• Addressing high-performance data analysis from massive simulation outputs.

Innovation

• Redesigning astrophysical codes for exascale systems
• Adopting innovative programming paradigms and software solutions
• Using machine learning for high-performance data analysis