Integrated Photonics for AI-Driven Data Centers
The explosive growth of artificial intelligence, machine learning, and large-scale data analytics is pushing traditional data center architectures to their physical and energy limits. As AI workloads become increasingly compute- and data-intensive, the need for faster, more scalable, and energy-efficient interconnect solutions has never been more urgent. In this context, integrated photonics, the use of light to move and process data on silicon chips, emerges as a groundbreaking technology that can revolutionize the way data centers operate.
This keynote will delve into how integrated photonics is poised to transform AI-driven data centers by enabling ultra-high bandwidth, low-latency communication with significantly reduced power consumption compared to conventional electronic systems. We will explore the fundamental principles of photonic integration, recent advances in optical interconnects, and the co-design of photonic systems with AI hardware. Special emphasis will be placed on how photonic technologies are addressing bottlenecks in data movement, inter-core communication, and thermal efficiency, critical challenges in modern computing environments.
In addition, the keynote will highlight real-world use cases, ongoing research from leading institutions such as MIT, and the industrial momentum behind deploying photonic chips at scale. Participants will gain a comprehensive understanding of the technical innovations, design trade-offs, and commercial implications of integrating photonics into next-generation data center infrastructure. The talk will also touch on future directions, including quantum photonics and AI-optics co-optimization, offering a vision of a more sustainable, high-performance computing ecosystem.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
- Contact Event Host
-
Prof. Mohammad Salah
Chair of IEEE IAS Jordan Chapter
Department of Mechatronics Engineering
Faculty of Engineering
E3130 Engineering Building
The Hashemite University
Office: (+962) 5 390-3333 Ext 4865 - Co-sponsored by IEEE IAS Jordan Chapter
Speakers
Dr. Anuradha Agarwal of Massachusetts Institute of Technology (MIT)
Integrated Photonics for AI-Driven Data Centers
The explosive growth of artificial intelligence, machine learning, and large-scale data analytics is pushing traditional data center architectures to their physical and energy limits. As AI workloads become increasingly compute- and data-intensive, the need for faster, more scalable, and energy-efficient interconnect solutions has never been more urgent. In this context, integrated photonics, the use of light to move and process data on silicon chips, emerges as a groundbreaking technology that can revolutionize the way data centers operate.
This keynote will delve into how integrated photonics is poised to transform AI-driven data centers by enabling ultra-high bandwidth, low-latency communication with significantly reduced power consumption compared to conventional electronic systems. We will explore the fundamental principles of photonic integration, recent advances in optical interconnects, and the co-design of photonic systems with AI hardware. Special emphasis will be placed on how photonic technologies are addressing bottlenecks in data movement, inter-core communication, and thermal efficiency, critical challenges in modern computing environments.
In addition, the keynote will highlight real-world use cases, ongoing research from leading institutions such as MIT, and the industrial momentum behind deploying photonic chips at scale. Participants will gain a comprehensive understanding of the technical innovations, design trade-offs, and commercial implications of integrating photonics into next-generation data center infrastructure. The talk will also touch on future directions, including quantum photonics and AI-optics co-optimization, offering a vision of a more sustainable, high-performance computing ecosystem.
Biography:
Dr. Anu Agarwal is a Principal Research Scientist at MIT, developing a Si-CMOS compatible platform for co-packaged optics and mid-IR sensing and imaging.
She is an Optica Fellow (2022), has over 250 journal and refereed conference publications, 21 awarded patents, and 1 pending patent.
She is director of electronic-photonic packaging at the MIT Microphotonics Center and leads the Lab for Education and Application Prototypes (LEAP) at MIT.nano within the Initiative for Knowledge and Initiative in Manufacturing (IKIM).
Through this LEAP initiative her team has built hybrid advanced manufacturing skills training programs to bridge the Knowledge, Skills, and Abilities gap in STEM across the workforce supply chain from K through Gray.
She leads a research program in resource-efficient microchip manufacturing and operations that seeks to build a global self-consistent roadmap across technology, value chain innovation, and workforce, for the semiconductor industry.
Website: https://photonics.mit.edu/people/principal-investigators/anu-agarwal
Email:
Address:Materials Research Laboratory, Massachusetts Institute of Technology (MIT), Cambridge, Massachusetts, United States