Neuromorphic systems open up opportunities to enlarge the explorative space for computational research. However, it is often challenging to unite efficiency and usability. This work presents the software aspects of this endeavor for the BrainScaleS-2 system, a hybrid accelerated neuromorphic hardware architecture based on physical modeling. We introduce key aspects of the BrainScaleS-2 Operating System: experiment workflow, API layering, software design, and platform operation. We present use cases to discuss and derive requirements for the software and showcase the implementation. The focus lies on novel system and software features such as multi-compartmental neurons, fast re-configuration for hardware-in-the-loop training, applications for the embedded processors, the non-spiking operation mode, interactive platform access, and sustainable hardware/software co-development. Finally, we discuss further developments in terms of hardware scale-up, system usability and efficiency.
@article{arxiv.2203.11102,
title = {A Scalable Approach to Modeling on Accelerated Neuromorphic Hardware},
author = {Eric Müller and Elias Arnold and Oliver Breitwieser and Milena Czierlinski and Arne Emmel and Jakob Kaiser and Christian Mauch and Sebastian Schmitt and Philipp Spilger and Raphael Stock and Yannik Stradmann and Johannes Weis and Andreas Baumbach and Sebastian Billaudelle and Benjamin Cramer and Falk Ebert and Julian Göltz and Joscha Ilmberger and Vitali Karasenko and Mitja Kleider and Aron Leibfried and Christian Pehle and Johannes Schemmel},
journal= {arXiv preprint arXiv:2203.11102},
year = {2022}
}