The speed of AI calculations doubles every 3.4 months, outpacing what standard semiconductor technologies are left to offer. Meanwhile, AI hardware is moving away from general-purpose applications in response to market demands that are increasingly verticalized by use cases. To accelerate the exponential progress of AI across industries, a new computing paradigm is now needed – one that is both faster and more application-specific.

Bridging this growing gap between the demand for AI processing and the supply of the Oxford-based semiconductor industry Salience Labs raised $ 11.5 million in funding to develop an ultra-high-speed multi-chip processor combining photonics and electronics to accelerate exponential progress in AI. The positional circle was led by Cambridge Innovation Capital and Oxford Science Enterprisess Oxford Investment Advisersformer CEO of Dialog Semiconductor Jalal Bagherleyformer member of the board of Temasek Yu Lin Goh and supported by the hand Deeptech Labs participating.

Salience Labs aims to commercialize an ultra-high-speed multi-chip processor that packs a photon chip along with standard electronics.

The company uses multi-chip design, with photon processing mapping directly on static random access memory (SRAM). This memory computing architecture is faster and more adaptable to the application-specific requirements of different market verticals, making it ideal for realizing AI applications in communications, robotics, vision, healthcare and other workloads. data.

Vaish Kewada, CEO and co-founder of Salience Labs, said: “The world needs faster and faster chips to develop AI capabilities, but the semiconductor industry can’t keep up with that demand. We address this interruption of supply and demand with our patented memory computing approach, which combines ultra-fast photonics, electronics flexibility and CMOS technology to deliver extra-scale computing to artificial intelligence applications. This will usher in a new era of processing, where supercomputer AI is becoming ubiquitous. “

Ian Lane, partner, Cambridge Innovation Capital said: “Salience Labs brings together deep expertise in photonics, electronics and CMOS manufacturing. Their unique approach to photonics provides an extremely dense computing chip without having to scale the photonic chip to large sizes.

https://tech.eu/2022/05/12/meet-salience-labs-oxford-startup-that-chipped-in-115-million-to-develop-high-speed-chips-for-ai-applications

Previous articleThe custom-made titanium OL + EM ebike does not cut any corners
Next articleTransforming the car supply chain for the 21st century