Chip startups using light instead of wires gain momentum and investment


April 26 (Reuters) – Computers using light rather than electric currents for processing, considered only a few years ago as research projects, are gaining momentum and startups that have solved the technical challenge of using photons in chips receive significant funding.

In the latest example, Ayar Labs, a startup developing the technology called silicon photonics, said on Tuesday it has raised $130 million from investors including chip giant Nvidia Corp (NVDA.O).

While the transistor-based silicon chip has increased computing power exponentially in recent decades as transistors have grown to the width of several atoms, shrinking them further is a challenge. Not only is it difficult to make something so tiny, but as they get smaller the signals can bleed between them.

Join now for FREE unlimited access to


Thus, Moore’s Law, which predicted that every two years the density of transistors on a chip would double and reduce costs, is slowing down, pushing the industry to seek new solutions to handle the increasingly heavy computing needs of the world. ‘artificial intelligence.

According to data firm PitchBook, silicon photonics startups raised more than $750 million last year, doubling from 2020. In 2016, that was around $18 million.

“AI is growing like crazy and taking over large parts of the data center,” Ayar Labs CEO Charles Wuischpard told Reuters in an interview. “The challenge of data movement and the power consumption in that data movement is a big, big problem.”

The challenge is that many large machine learning algorithms may use hundreds or thousands of chips for computation, and there is a bottleneck on the speed of data transmission between chips or servers using electrical methods. current.

Light has been used to transmit data over fiber optic cables, including undersea cables, for decades, but getting it down to the chip level was difficult because the devices used to create the light or the control weren’t as easy to shrink as transistors.

PitchBook Senior Emerging Technologies Analyst Brendan Burke expects silicon photonics to become mainstream hardware in data centers by 2025 and estimates the market will reach $3 billion by then , which is similar to the size of the AI ​​graphics chip market in 2020.

Beyond connecting transistor chips, startups using silicon photonics to build quantum computers, supercomputers and chips for autonomous vehicles are also raising significant funds.

PsiQuantum has raised around $665 million so far, though the promise of quantum computers changing the world is still a long way off.

Lightmatter, which builds processors using light to accelerate AI workloads in the data center, has raised a total of $113 million and will release its chips later this year and test them with customers shortly. after.

Luminous Computing, a startup building an AI supercomputer using silicon photonics backed by Bill Gates, has raised a total of $115 million.

It’s not just startups pushing this technology forward. Semiconductor makers are also gearing up to use their silicon chip fabrication technology for photonics.

Amir Faintuch, head of computing and wired infrastructure at GlobalFoundries, said the collaboration with PsiQuantum, Ayar and Lightmatter has helped create a silicon photonics fabrication platform that others can use. The platform was launched in March.

Peter Barrett, founder of venture capital firm Playground Global, investor in Ayar Labs and PsiQuantum, believes in the long-term prospects of silicon photonics to speed up computation, but says there is a long way to go.

“What the guys at Ayar Labs are doing so well…is they’ve solved the data interconnect problem for traditional high performance (computing),” he said. “But it will be some time before we have a pure digital photonic computation for non-quantum systems.”

Join now for FREE unlimited access to


Reporting by Jane Lanhee Lee; Editing by Stephen Coates

Our standards: The Thomson Reuters Trust Principles.


About Author

Comments are closed.