The largest SpiNNaker2 supercomputer for molecule drug discovery yet, with 656,640 cores
DRESDEN, GERMANY / ACCESS Newswire / July 28, 2025 / Today SpiNNcloud secured a deal to deliver a brain-inspired supercomputer to Leipzig University, marking a significant milestone in the advancement of computing for small molecule research for personalized medicine. This is the largest SpiNNcloud System to be deployed specifically for drug discovery, encompassing around 650k cores.
Developed by deep-tech company SpiNNcloud, SpiNNcloud Server System can simulate up to a minimum of 10.5 billion neurons for AI, HPC, and other applications. The SpiNNcloud Server System is based on the current chip generation and was pioneered by Steve Furber, designer of the original ARM architecture,and uses a large number of low-power processors for efficiently computing AI and other workloads.
The multi-million Euro system to be deployed in Leipzig is intended for protein folding simulations to achieve personalized medicine. The approach leverages the extreme parallelism and scale of these systems to deploy millions of small models tasked with finding hits between molecules and patient profiles. The system will enable the discovery of new personalized drugs at faster convergence speeds and under a lower energy profile compared to traditional GPU-based systems.
“The SpiNNcloud Server System architecture makes screening billions of molecules in silico feasible with a brain-inspired supercomputer design,” said Christian Mayr, SpiNNcloud co-founder. “Originally dedicated to biological neural network simulation, The SpiNNcloud Server System is tailored for massively parallel execution of small, heterogeneous compute workloads, with generally programmable 10 million ARM-based processors with many dedicated DNN accelerators. A prototype neural network allows the screening of 20 billion molecules in less than an hour, two orders of magnitude faster than on 1000 CPU cores.”
“The extreme parallelism of SpiNNcloud supercomputers makes them ideally suited for protein folding applications like those used in small molecule drug discovery,” said Jens Meiler, Alexander von Humboldt Professor and Director of the Institute for Drug Discovery at Leipzig University. “Protein folding can be viewed as an optimization problem where the protein aims to find its lowest energy state. SpiNNcloud supercomputers are good at solving optimization problems and can be used to find optimal protein structures.”
The SpiNNcloud system employs a highly parallel architecture consisting of 48 SpiNNaker2 chips per server board, each containing 152 Arm-based cores with specialized accelerators. This design enables efficient, event-driven computation, allowing the system to perform complex simulations at a lower energy profile compared to traditional GPU-based systems. Such energy efficiency is crucial for applications where power consumption and cooling are limiting factors.
“Our brain-inspired computing architecture is uniquely suited for deploying efficient algorithms that require dynamic sparsity and extreme parallelism,” said Hector Gonzalez, SpiNNcloud co-founder and CEO. “Our systems are 18 times more energy efficient than current GPUs and are being used by leading institutions across Europe and the US. The deployment in Leipzig demonstrates the flexibility of our systems, as well as the continued adoption of the technology for unrivaled performance and energy efficiency.”
Looking ahead from a broader perspective, SpiNNcloud is enabling support for the next generation of Gen AI algorithms, paving a radically more efficient path to machine learning advancement through dynamic sparsity. Recent breakthroughs in machine learning are driving a transition from traditional dense modeling – centered on fixed feature selection within neural representations – to extreme dynamic sparsity, where a subset of neural pathways are selectively activated based on the input. This approach helps to shape entirely new architectures for AI foundation models, and addresses the current energy crisis driven by mainstream AI scaling trends.
“SpiNNcloud’s approach reflects a broader shift in performance-intensive computing, where innovation demands that infrastructure and algorithms be co-designed from the ground up,” said Peter Rutten, Research Vice-President, Performance Intensive Computing, Worldwide Infrastructure Research at IDC.
For more information about SpiNNcloud and the SpiNNcloud platform, please visit https://spinncloud.com.
About SpiNNcloud:
SpiNNcloud enables customers to realize complex AI-driven systems through brain-inspired supercomputing technology. Our SpiNNaker2 architecture supports dynamic sparsity in mainstream AI applications while delivering superior energy efficiency. As a winner of the prestigious EIC Accelerator program, SpiNNcloud continues to solidify itself as sole European champion for specialized AI Supercomputers.
Contact:
Forrest Carman
Owen Media
[email protected]
SOURCE: SpiNNcloud Systems GmbH
