According to Reuters, GlobalFoundries has just announced its acquisition of Singapore-based silicon photonics device manufacturer AMF, which the company says makes it the largest manufacturer in the technology field. 

Silicon photonics devices use light instead of electrical pulses to transmit data and can be used for communication within chips, between components, and even between servers.

Silicon photonics technology

Nvidia plans to incorporate silicon photonics technology in its next-generation AI servers, which will reduce power consumption while simultaneously increasing data transfer speeds. This is crucial for the company, as it enables clusters of millions of GPUs within data centers. AMD is also actively developing this technology, reportedly investing nearly $300 million to establish a dedicated R&D center in Taiwan. Earlier this year, AMD acquired another silicon photonics company, Enosemi, to enhance its competitiveness against Nvidia.

Silicon photonics technology's ability to increase data transfer speeds while reducing power consumption is critical for the future of AI computing. This is especially true given the unprecedented expansion of AI data centers and the resulting surge in electricity demand, placing immense pressure on the power grid. 

"As data transfer speeds increase and workloads become more complex, the ability to transmit information with greater speed, accuracy, and energy efficiency is essential for AI data centers and advanced telecommunications networks," said Tim Breen, CEO of GlobalFoundries, in a statement.

Quantum computing

Beyond its applications in artificial intelligence, silicon photonics is also a key technology for quantum computing. Using optical signals instead of electrical signals allows for the creation of systems that do not require cryogenic cooling, making quantum computers more practical and cost-effective.

While giants like AMD, Nvidia, and GlobalFoundries are investing in silicon photonics technology, other startups are also joining the race. Companies like Ayar Labs, Celestial AI, and Lightmatter are developing their own technologies and photonics-based chips, especially as photonics is seen as the future of computing.

While we don't expect consumer-grade CPUs, GPUs, and motherboards to adopt this technology in the near future, it is already being used in large servers that process terabytes of data per second. With continued investment in AI data centers, we anticipate further progress in the research and development of this promising technology.

Source: Content compiled from tomshardware

View more at EASELINK