GlobalFoundries believes data traveling at the speed of light is the future, and it's pointing to support from Nvidia, Cisco Systems, and others as evidence that its silicon photonics manufacturing tech is mainstream-ready.

The Malta, NY-based chip manufacturer said it has designs under way with major customers for its new GF Fotonix platform. Announced Monday, the new platform will address ever-growing needs to process data faster using less energy while also maintaining the company's leadership in the fast-growing optical networking module market, according to the company.

"Silicon photonics is now recognized as a necessary technology for the datacenter revolution, and our leading semiconductor manufacturing technology platform accelerates adoption into the mainstream," said Amir Faintuch, senior vice president and general manager of Compute and Wired Infrastructure at GlobalFoundries.

[1]

GF Fotonix is a monolithic platform, bringing together, for the first time in the industry, 300mm photonics features and a 300GHz-class RF-CMOS on a silicon wafer, GlobalFoundries said. These photonics, radio frequency, and complementary metal–oxide–semiconductor components were previously manufactured onto separate chips that had to be combined in a single chip.

[2]

[3]

The benefits of moving to a monolithic design are significant, according to GlobalFoundries. The company said GF Fotonix has shown to deliver 0.5Tbps/fiber, the highest data rate per fiber in the industry. This will allow the company to manufacture optical chiplets with speeds of 1.6-3.2Tbps, providing faster and more efficient data transmission and better signal integrity.

GlobalFoundries believes GF Fotonix has an edge for next-generation AI applications, thanks to its 10,000x improvement in system error rate.

[4]

The firm also pointed to GF Fotonix's high level of integration on a photonics integrated circuit. The company claimed it will allow customers to pack more product functions on a chip and "simplify their bill of materials."

GF Fotonix supports various chip packaging solutions too, including a passive attachment for larger fiber arrays, 2.5D packaging and on-die lasers.

GlobalFoundries said the first version of its Process Design Kit for GF Fotonix will be available in April, which will be complemented with design tools from Ansys, Cadence Design Systems, and Synopsis.

[5]PsiQuantum envisions its data-center-sized quantum computer

[6]Photonics startup Luminous Computing bags $105m

[7]Machine-learning model pinpoints dying power grid components

[8]Top chipmakers ignore India's semiconductor factory subsidies

The chip manufacturer is hoping to make a big splash with GF Fotonix after [9]going public last year , and it has lined up testimonials from multiple companies that plan to use the platform.

Nvidia expects a boost in high-performance computing and AI applications for the "leading-edge" data center products that will use high-bandwidth, low-power optical interconnects designed using the GF Fotonix platform, according to a statement provided by GlobalFoundries.

[10]

As for Cisco, GlobalFoundries has partnered with the networking giant on a custom silicon photonics solution for data center network and data center interconnect applications.

"Our heavy investment and leadership in silicon photonics, combined with GF's feature-rich manufacturing technology, allows us to deliver best-in-class products," Bill Gartner, head of Cisco's Optical Systems and Optics Group, said in a statement.

Other companies noted for using GF Fotonix include quantum startup PsiQuantum and chip startup Ayar Labs. GlobalFoundries also provided statements in support of its broader silicon photonics efforts from Marvell, Broadcom, Lightmatter, and Macom. ®

Get our [11]Tech Resources



[1] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2YieLuHujA6slDSqA61CafgAAAMs&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0

[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YieLuHujA6slDSqA61CafgAAAMs&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0

[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33YieLuHujA6slDSqA61CafgAAAMs&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0

[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YieLuHujA6slDSqA61CafgAAAMs&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0

[5] https://www.theregister.com/2022/03/08/psiquantum_data_center_quantum/

[6] https://www.theregister.com/2022/03/05/luminous-ai-supercomputer-photonics/

[7] https://www.theregister.com/2022/02/26/machine_learning_power/

[8] https://www.theregister.com/2022/02/21/india_semiconductor_subsidies/

[9] https://www.theregister.com/2021/10/04/globalfoundries_files_ipo_us/

[10] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33YieLuHujA6slDSqA61CafgAAAMs&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0

[11] https://whitepapers.theregister.com/