Next-Generation AI Hardware: 3D Photonic Electronic Platform Increases Efficiency and Bandwidth

Artificial intelligence (AI) systems promise transformative advancements, but their growth is limited by energy inefficiencies and bottlenecks in data transfer. Researchers at Columbia Engineering have announced a groundbreaking solution. Pleasing the path to next-generation AI hardware with a 3D photonic electronic platform that achieves unprecedented energy efficiency and bandwidth density.

The study, “3D Photonics for Ultra Low Energy, High Bandwidth Chip Data Links,” led by Kellenbergman, professor of electrical engineering at Charles Bachelor, is featured in Nature Photonics.

This study details a pioneering way to integrate photonics with sophisticated complementary oxide-semiconductor (CMOS) electronics to redefine energy-efficient high-bandwidth data communications. This innovation addresses key challenges in data movement. This is a persistent obstacle to achieving faster, more efficient AI technology.

“This work presents technology that can transfer huge amounts of data with unprecedented energy consumption,” Bergman said. “This innovation breaks through long-standing energy barriers that limit the movement of data in traditional computer and AI systems.”

Data communication breakthrough

The Columbia Engineering team collaborated with professors of Alyosha Christopher Molner, Ilda and Charlesley Engineering at Cornell University to develop a 3D integrated photonic electronic chip with a high density of 80 photonic transmitters and receivers within a compact chip footprint.

The platform offers high bandwidth (800 GB/s) with exceptional energy efficiency and consumes 120 femtojoules per bit. The bandwidth density is 5.3 Tb/s/mm2, and this innovation is well beyond existing benchmarks.

Designed for low cost, the chip integrates photonic devices with CMOS electronic circuits, leveraged components manufactured in commercial foundry, and sets stages of adoption for a wide range of industry.

AI hardware innovation

The team’s research redefines how data is transmitted between computing nodes and addresses long-standing bottlenecks in energy efficiency and scalability.

With 3D integrated photonic and electronic chips, this technology delivers unparalleled energy savings and high bandwidth density, freeing it from the constraints of traditional data locality. This innovative platform allows AI systems to efficiently transfer huge amounts of data and support previously impractical distributed architectures due to energy and latency limitations.

The resulting advances are poised to unlock unprecedented levels of performance, and the technology is the basis for future computing systems across applications, from large-scale AI models to real-time data processing of autonomous systems.

Beyond AI, this approach has the potential to transform high-performance computing, communications, and decomposed memory systems, marking a new era of energy-efficient high-speed computing infrastructure.

The collaborative research included contributions from Cornell University’s Molner Institute, Air Force Institute, and Dartmouth University.