Intel has recently announced the creation of Hala Point, the world’s largest neuromorphic system, marking a major step towards more sustainable and efficient artificial intelligence. Deployed initially at Sandia National Laboratories, Hala Point uses Intel’s advanced Loihi 2 processor and builds on the success of its predecessor, Pohoiki Springs, by offering substantial improvements in architecture. This enhancement boosts neuron capability by greater than tenfold and performance by as much as twelve times.
“The computing cost of today’s AI models is rising at unsustainable rates. The industry needs fundamentally latest approaches able to scaling. For that reason, we developed Hala Point, which mixes deep learning efficiency with novel brain-inspired learning and optimization capabilities. We hope that research with Hala Point will advance the efficiency and adaptableness of large-scale AI technology,” said Mike Davies, director of the Neuromorphic Computing Lab at Intel Labs.
Hala Point distinguishes itself by being the primary large-scale neuromorphic system able to demonstrating state-of-the-art computational efficiencies on mainstream AI workloads. It may well support as much as 20 quadrillion operations per second, or 20 petaops, and offers unprecedented efficiency exceeding 15 trillion 8-bit operations per second per watt (TOPS/W) when executing conventional deep neural networks.
Researchers at Sandia National Laboratories will use Hala Point for advanced brain-scale computing research, specializing in scientific computing problems across various domains corresponding to device physics, computer architecture, and informatics. “Working with Hala Point improves our Sandia team’s capability to resolve computational and scientific modeling problems. Conducting research with a system of this size will allow us to maintain pace with AI’s evolution in fields starting from business to defense to basic science,” stated Craig Vineyard, Hala Point team lead at Sandia National Laboratories.
While Hala Point stays a research prototype, Intel envisions its lessons will significantly enhance future business systems’ capabilities, notably enabling large language models to learn repeatedly from latest data and reducing the training burden of AI deployments.
The drive for increasingly large deep learning models has exposed significant sustainability challenges inside AI, necessitating innovation at the basic levels of hardware architecture. Neuromorphic computing, inspired by neuroscience, integrates memory and computing inside a highly parallel framework to reduce data movement. This approach has demonstrated remarkable gains in efficiency, speed, and adaptableness, as evidenced by Loihi 2’s performance at this month’s International Conference on Acoustics, Speech, and Signal Processing (ICASSP).
Hala Point integrates 1,152 Loihi 2 processors and supports as much as 1.15 billion neurons and 128 billion synapses, distributed over 140,544 neuromorphic processing cores, inside a six-rack-unit data center chassis. Its massively parallelized fabric offers significant memory bandwidth and communication speeds, providing a strong foundation for bio-inspired spiking neural network models.
Intel’s ongoing development of neuromorphic systems like Hala Point goals to handle power and latency constraints that currently limit the real-world deployment of AI. With the continued collaboration of the Intel Neuromorphic Research Community (INRC), Intel is committed to advancing this brain-inspired technology from research prototypes to business products.
