SpiNNcloud Systems introduces Neuromorphic Supercomputer for Commercial Use

SpiNNcloud Systems has made an exciting announcement today, unveiling its highly anticipated SpiNNaker2 platform ahead of ISC High Performance 2024. This supercomputer-level hybrid AI high-performance computer system is built on the principles of the human brain, marking a significant milestone in the field of artificial intelligence.

The SpiNNaker2 platform is the brainchild of Steve Furber, for his work on the original ARM and SpiNNaker1 architectures. Leveraging a multitude of low-power processors, this system offers efficient computing for AI and other workloads. The first-generation SpiNNaker1 architecture has already gained widespread adoption, with research groups in 23 countries utilizing its capabilities. Notable customers such as Sandia National Laboratories, Technical University of München, and Universität Göttingen have already placed orders for SpiNNaker2.

The development of SpiNNaker2 was made possible through the commercialization of intellectual property from the Human Brain Project, a billion-euro research initiative funded by the European Union. Designed to create intelligent and efficient artificial systems, this project has paved the way for brain-like computation in the realm of AI.

Fred Rothganger from Sandia National Labs expressed his enthusiasm for SpiNNaker2, highlighting its flexibility as a neural supercomputer architecture. “SpiNNaker2 is the most flexible neural supercomputer architecture available today,” he said. “At Sandia, we are excited to build applications on this amazing system.”

So what sets SpiNNaker2 apart? Let’s delve into its technical specifications. The SpiNNcloud server board comprises 48 SpiNNaker2 chips, each housing a low-power mesh of 152 Arm-based cores. These cores are equipped with various accelerators, including distributed GPU-like units that enhance neuromorphic, hybrid, and mainstream AI model computations.

One of SpiNNaker2’s key strengths lies in its hybrid capabilities, which allow it to combine practical algorithms from machine learning, energy efficiency from event-based operation, and the reliability of rule-based layers governing application backends. This unique combination empowers SpiNNaker2 to deliver exceptional performance across a range of workloads.

Designed for scalability, SpiNNaker2 can be expanded through a multi-rack setup, with each rack comprising 90 SpiNNcloud server boards. This configuration enables the system to support a staggering 10 billion interconnected neurons firing in real-time. Additionally, as a pre-exascale supercomputer capable of allocating machine learning workloads, SpiNNaker2 can deliver up to 0.3 exaops of performance.

SpiNNaker2 distinguishes itself from GPU solutions by offering greater versatility, improved performance, lower operational costs, and better availability. By utilizing numerous asynchronous low-power units, SpiNNaker2 tackles workloads more efficiently, resulting in energy-proportional operation, enhanced communication, and reduced operational expenses. This neuromorphic supercomputing architecture sets a new standard in the field.

But SpiNNaker2 doesn’t stop at neuromorphic architecture; it also offers hybrid AI acceleration, a crucial aspect in designing systems that fall under DARPA’s definition of the third wave of AI. These systems possess an understanding of their context and environment and develop explanatory models over time to characterize real-world phenomena.

SpiNNcloud Systems plans to initially offer SpiNNaker2 as a cloud platform solution, allowing early adopters to access this technology cost-effectively. Co-founder and co-CEO Hector Gonzalez shares the company’s vision for the future: “Our vision is to pioneer the future of artificial intelligence through brain-inspired supercomputer technology and to create systems that are not only powerful but also transparent and explainable.”

The cloud service will be available in the second half of 2024, with full production systems ready for shipment in the first half of 2025. SpiNNaker2’s market-leading position in hybrid AI high-performance computing opens up exciting possibilities in defense, drug discovery, quantum emulation, smart city applications, and more. With its advanced brain-like supercomputing platform, SpiNNcloud Systems is poised to shape the future of artificial intelligence.

Background Information

---

About ARM:

ARM, originally known as Acorn RISC Machine, is a British semiconductor and software design company that specializes in creating energy-efficient microprocessors, system-on-chip (SoC) designs, and related technologies. Founded in 1990, ARM has become a important player in the global semiconductor industry and is widely recognized for its contributions to mobile computing, embedded systems, and Internet of Things (IoT) devices. ARM's microprocessor designs are based on the Reduced Instruction Set Computing (RISC) architecture, which prioritizes simplicity and efficiency in instruction execution. This approach has enabled ARM to produce highly efficient and power-saving processors that are used in a vast array of devices, ranging from smartphones and tablets to IoT devices, smart TVs, and more. The company does not manufacture its own chips but licenses its processor designs and intellectual property to a wide range of manufacturers, including Qualcomm, Apple, Samsung, and NVIDIA, who then integrate ARM's technology into their own SoCs. This licensing model has contributed to ARM's widespread adoption and influence across various industries.

  

Latest Articles about ARM

Technology Explained

---

GPU: GPU stands for Graphics Processing Unit and is a specialized type of processor designed to handle graphics-intensive tasks. It is used in the computer industry to render images, videos, and 3D graphics. GPUs are used in gaming consoles, PCs, and mobile devices to provide a smooth and immersive gaming experience. They are also used in the medical field to create 3D models of organs and tissues, and in the automotive industry to create virtual prototypes of cars. GPUs are also used in the field of artificial intelligence to process large amounts of data and create complex models. GPUs are becoming increasingly important in the computer industry as they are able to process large amounts of data quickly and efficiently.

Latest Articles about GPU