Realistic check of AI interconnect scaling solutions for post 1.6T era

Time: Monday, 2 November 2026 | 1:30 p.m. - 5:00 p.m.

Location: Conference Room 205, 2F

 

Session Description:

With AI DC build out for 2026, rapid deployment of 800G is happening with close to 100M ports annually, while 1.6T is emerging soon as clear roadmap to reach prime time in next 2-3 years. The next era of networking is here for post 1.6Tb/s. In order to sustain the AI and cloud‑driven bandwidth surge, the industry enters into another crossroad again with many MSA solutions available, similar to the last situation with transition to 100Gb/s for 10 year ago. From industry perspective, while CPO is commonly considered as a long‑term enabler, many other solutions like NPO, CPX, XPO, 300/400G per lane, high‑order modulation & Coherent lite are proposed. In this session, we will provide an overview of various development status/readiness, and discuss these high capacity/high density optical solutions per use cases for how to meet AI scaling requirements in supporting AI scale up, scale out and scale across. 


More specially, the next revolution in AI computing clusters has been driven by scale up (for vertical scaling) as botleneck to enhance single-node performance such like “Super nodes”.  As we race toward 1.6T and beyond, it pushes network infrastructure to its fundamental limits of physical electrical interconnects, so a new paradigm is emerging for post 1.6T sotions:: a fundamental shift in how industry design network hardwares beyond pluggable optics transceivers. 


The workshop will focus on hyperscale end user case studies, design/package challenges, implementation & commercialization. Looking beyond today’s SuperPods, the path toward increasingly large and tightly connected “Hyper Pods,” become prevailing  with optics enabling  next level of system scale and integration. The workshop will bring together hyperscalers, CSP end users, switch integrators, and technology innovative vendors to exchange technology and market intelligence and help align future products, roadmaps, and supply chains around the next generation of AI computing.


Keywords: Scale up, 1.6T, optical transceivers, Switches, Data Center Interconnect


Organisers:

Scott Wilkinson, Cignal AI

Frank Chang, Source Photonics

Guangcan Mi, Huawei


Speakers:

Mark Lutkowitz, fibeReality, LLC, USA

Title: Pragmatic Evaluation of Existing and New Solutions Beyond 1.6T

Eric Berner, Huawei, Canada

Title: Scaling Al interconnect: optical innovation in NPO solutions

Chongjin Xie, PhotonicX AI, Singapore

Title: Al optical interconnects beyond 1.6T: pluggable, NPO & CPO

Marcus Yang, SiPhi, USA

Title: CPO Is Only the Beginning: What Optical Interconnect Does AI Datacenter Need?

Samuel Liu, Lumilens, USA

Title: Will NPO play a major role for post 1.6T AL optical interconnect applications?

Break for 15min

Mingyang lyu, Senior principal analyst, Omdia, USA

Title: Geopolitical Impacts and Resilience in Optical Component Manufacturing

Minqing Zuo, China Mobile, China

Title: AI-Computing-Oriented DCI: System and Optical Fiber

Anbin Wang, Alibaba, China

Title: Al-driven Optical interconnect beyond 1.6T

Die Wan, Baidu, China

Title: Scale-up-Driven Optical Interconnect Transformation

Son Thai Le, Ciena, USA

Title: NPO/CPO Optical Engine for Al Scale-Up and Scale-Out?

Xiang Liu, Huawei, Hong Kong, China

Title: Scaling and Standardization of Al Interconnect Solutions


HOSTS
WESTLAKE UNIVERSITY

WESTLAKE UNIVERSITY

ZHEJIANG UNIVERSITY

ZHEJIANG UNIVERSITY

CO-ORGANIZERS
POC

POC

LUSTER

LUSTER

TECHNICAL SPONSORS
IEEE Photonics Soceity

IEEE Photonics Soceity

OPTICA

OPTICA

SPIE

SPIE

THE CHINESE OPTICAL SOCIETY

THE CHINESE OPTICAL SOCIETY

CHINA INSTITUTE OF COMMUNICATIONS

CHINA INSTITUTE OF COMMUNICATIONS

CHINA INSTITUTE OF ELECTRONICS

CHINA INSTITUTE OF ELECTRONICS

CONFERENCE SUPPORT
Learning Conference

Learning Conference

BRIGHTEN CORE

BRIGHTEN CORE