Intel Launches World’s Largest Neuromorphic System For More Efficient And Sustainable AI

he global demand for powerful artificial intelligence (AI) capabilities has led AI companies and technology conglomerates to shift into overdrive in a bid to design, develop and deploy the most advanced products, services, software applications and infrastructure.

Looking to stay ahead of the pile, Intel has come up with an announcement that’s set to boost neuromorphic computing to beat more conventional architectures and boost AI learning capabilities.
So, what did Intel reveal? Let’s explore!
 

What Did Intel Announce?

 

• In a news release published on its website, Intel announced that it has built the world’s largest neuromorphic system, code-named Hala Point.

• Neuromorphic systems are designs inspired by the human brain's structure and operation and use artificial neurons and synapses to carry out computations.

• The Hala Point system, which is a prototype, uses 1,152 Loihi 2 processors arranged in a six-rack-unit data center chassis the size of a microwave oven. The system also consists of 2,300+ embedded x86 processors for ancillary computations.

• As per the release, “The system supports up to 1.15 billion neurons and 128 billion synapses distributed over 140,544 neuromorphic processing cores.”

• Intel’s Hala Point furthers the company’s first-generation large-scale research system, Pohoiki Spring, using architectural improvements that help it achieve 10x more neuron capacity and 12x higher performance.

• As per the release, “Hala Point is the first large-scale neuromorphic system to demonstrate state-of-the-art computational efficiencies on mainstream AI workloads,” that “can support up to 20 quadrillion operations per second, or 20 petaops, with an efficiency exceeding 15 trillion 8-bit operations per second per watt (TOPS/W) when executing conventional deep neural networks.”

• Essentially it can surpass architectures built on graphics processing units (GPU) and central processing units (CPU) enabling future real-time continuous learning for AI applications.

• The system was initially deployed at Sandia National Laboratories, where it will be used for advanced brain-scale computing research.

• As per the release, “Hala Point integrates processing, memory, and communication channels in a massively parallelized fabric ... can process over 380 trillion 8-bit synapses and over 240 trillion neuron operations per second ... can execute its full capacity of 1.15 billion neurons 20 times faster than a human brain and up to 200 times faster rates at lower capacity.”

 

What Did Stakeholders Say?

 

• Mike Davies, Director of the Neuromorphic Computing Lab at Intel Labs, said, “The computing cost of today’s AI models is rising at unsustainable rates. The industry needs fundamentally new approaches capable of scaling.”

• [Contd.] “For that reason, we developed Hala Point, which combines deep learning efficiency with novel brain-inspired learning and optimization capabilities. We hope that research with Hala Point will advance the efficiency and adaptability of large-scale AI technology.”

• Craig Vineyard, Hala Point team lead at Sandia National Laboratories, said, “Working with Hala Point improves our Sandia team’s capability to solve computational and scientific modeling problems. Conducting research with a system of this size will allow us to keep pace with AI’s evolution in fields ranging from commercial to defense to basic science.”

Do you think this move will help Intel boost its position in the AI industry? Do you think Intel’s competitors need to make similar moves?

Let us know in the comments below!