Schedule

Heterogeneous Integration opens new capabilities for technologies, use cases, and market participants. Integrating devices developed separately and often on different substrates (as, for example, III-V semiconductors and Silicon Photonics) into a single composite device enables miniaturization, simplifies interconnection, saves materials, and reduces energy consumption. It fosters the creation of new optical products that serve new applications with more demanding performance requirements. In this session, we explore how well (and when) the various approaches to heterogeneous integration will meet these challenging needs., and the status and challenges for the service providers (e.g., type of integration platforms, process/tools maturity, and how to handle non-traditional Si materials).

Organizers

• Lukas Chrostowski, University of British Columbia, Canada
• Patrick Lo, Advanced Micro Foundry, Singapore
• Dan Pitt, Palo Alto Innovation Advisors, USA
• Mengjie Yu, University of Southern California, USA

Speakers

• Alex  Chikhaoui, X-Celeprint, Ireland
• Robert Kalman, Avicena, USA
• Michael Lebby, Lightwave Logic, USA
• Sylvie Menezo, Scintil, France
• Edward Preisler, Tower Semiconductor, USA
• Günther Roelkens, Ghent University, Belgium
• Mian Zhang, Hyperlight, USA

The application of optical switching in data center networks has been extensively studied. Google’s recent announcement showcasing the implementation of an optical circuit switch in a production data center has re-sparked interest in the requirements and challenges associated with optical switching. In addition, generative AI models are extensively advancing, with the number of parameters exponentially increasing. This requires GPU clusters with a very high bandwidth density and low energy consumption, which has revamped research on photonic switch fabrics co-integrated and optically interconnected with multiple GPUs/TPUs/CPUs. This workshop discusses the challenges and opportunities of optical switching for large-scale data center networks, especially for GPU clusters and HPC networks, from a system networking perspective (capacity demand, latency, fast configuration, and control scheme, flexibility and scalability, cost, power consumption), optical switch architectures (switch radix, topology, size and scale, performance), to device performance (loss, bandwidth, switching speed, crosstalk, and integration platform). We will explore innovative photonic technologies and network architectures for enabling optical switching for AI/ML applications. Some of the topics that we intend to dive into in this workshop are:

1. What are the requirements and challenges for the broad adoption of optical circuit switching?

2. How do optical and electrical switch systems co-exist to enable scaling and optimize the cost-to-performance metric for AI/ML systems?

3. Will semiconductor-based optical switches (e.g., Silicon Photonics) play a significant role in large-scale integrated optical switches after the MEMS-based OCS systems deployment? What hurdles need to be overcome?

4. Is there a role for fast optical switching in AI/ML systems? What are the requirements for fast optical switching?

5. Novel system architectures and packaging techniques involving co-packaged optics, optical interposers, and I/Os, with GPU/CPU/TPU for future AI/ML computing?

Organizers

• Qixiang Cheng, Cambridge University, UnitedKingdom
• Wenhua Lin, Intel Corp., USA
• Kazuhiro  Ikeda, AIST, Japan

Speakers

• Keren Bergman, Columbia University, USA
• Darius Bunandar, Lightmatter, USA
• Ben Lee, NVIDIA, USA
• Shu Namiki, AIST, Japan
• George Papen, UCSD, USA
• Richard Penty, Cambridge University, UnitedKingdom
• Stefano Stracca, Ericsson, Italy
• Ming Wu, University of California, Berkeley, USA

Started in 2023, OFCnet brings a new opportunity to the exhibition and demonstrates products, concepts, solutions, research, and architectures in live high-speed optical networks connected to the leading research and education networks worldwide. This increased focus on designing and building next-generation Optical Networks will expand exposure to connectivity, emerging and next-generation network technologies such as Quantum Networks, programmable and software-defined optical networks, and their applications such as big data, security, and distributed classical and quantum computing. This workshop brings together the innovators and researchers who work on the mentioned topics to enrich the OFCnet community further and expand the contributing parties. We discuss how this initiative should be developed to ensure OFCnet enriches future community participation.

Organizers

• Gwen Amice, EXFO, Canada
• Cees de Laat, University of Amsterdam, Netherlands
• Andrew  Lord, British Telecom, UnitedKingdom
• Reza  Nejabati, University of Bristol, UnitedKingdom

Speakers

• Chris Janson, Nokia, USA
• Joe Mambretti, Northwestern University, USA
• Corey McClelland, Qubitekk, USA
• Yoshiaki Sone, NTT, Japan