Unifying von-Neumann HPC and Neuromorphic Acceleration via the EBRAINS Research Infrastructure: A Framework for High-Performance Workflows

2026-06-07

AI Summary

This work addresses the lack of a unified execution framework for scientific workflows spanning traditional von Neumann supercomputers and neuromorphic hardware, which has led to fragmented computational pipelines. The authors present a cloud-native workflow system built on EBRAINS JupyterLab that leverages federated identity management to orchestrate tasks across conventional supercomputing platforms (JUSUF and Galileo100) and the SpiNNaker-1 neuromorphic system. For the first time, this approach enables model-level portability of spiking neural networks. By integrating zero-install Apptainer containers, a PMIx-aware runtime, and the domain-specific language NESTML, the system supports “write once, auto-compile” deployment to either NEST or sPyNNaker backends. Validation using a balanced random network benchmark demonstrates significantly improved execution consistency and reproducibility across heterogeneous computing infrastructures.

Abstract

Modern scientific workflows increasingly span diverse computing architectures, yet executing a single computational model across disparate systems often forces researchers to maintain fragmented, site-specific pipelines. In this paper, we address this challenge within the domain of computational neuroscience by presenting a unified, cloud-based workflow orchestrated via EBRAINS JupyterLab. This workflow enables users to transparently execute spiking neural networks on both von-Neumann supercomputers and neuromorphic hardware. Using a single federated identity, the system dispatches jobs to HPC sites (JUSUF, Galileo100) via PyUNICORE and to the SpiNNaker-1 neuromorphic system via the Neuromorphic Computing Platform Interface. To guarantee cross-site reproducibility and mitigate software version drift, we utilize a zero-installation execution mode that dynamically pulls PMIx-aware Apptainer containers to HPC compute nodes. Furthermore, we demonstrate genuine model-level portability using the NESTML domain-specific language, allowing custom neuron models to be written once and automatically compiled for either the NEST (C++) or sPyNNaker backends. Validated with a balanced random network case study, this work illustrates a practical, end-to-end path for hardware-agnostic workflows while highlighting the critical role of containerization and domain-specific languages in achieving true cross-platform reproducibility.

Problem

• scientific workflows
• cross-platform reproducibility
• computational neuroscience
• heterogeneous computing
• model portability

Innovation

• neuromorphic computing
• high-performance computing
• containerization
• domain-specific language
• hardware-agnostic workflow