Program

HOTI 2025 proceedings Get early access to HotI Proceedings via QR code or link


Announcement NOTE: Q&A's are handled in Slack, the link is provided in the Zoom Lobby!


The times in the program below are Pacific Time (PT).

Conference Day 1: Wednesday, August 20

Time Title Presenters
09:20 Introduction and Day 1 Welcome General Chair: Artem Polyakov (NVIDIA)
09:30 - 10:30 Keynote 1: Now in Focus: the Fifth, GenAI Epoch of Computing Infrastructure
[details]

Session chair: Artem Polyakov (NVIDIA)
Amin Vahdat (Google)
10:30 - 10:45 Diamond Sponsor Talk: Lightmatter
Transforming AI Infrastructure with the Power of 3D Photonics [details] Download PDF
Nick Harris (Lightmatter)
10:45 - 11:00 Diamond Sponsor Talk: Cornelis
Reimagining the Scale-Out Architecture for HPC and AI [details] Download PDF
Matt Williams (Cornelis Networks)
11:00 - 12:00 Technical Paper Session A:
Network Synchronization and Routing
Session chair: Atiqul Mollah (Cornelis Networks)
bittide: Control Time, Not Flows Download PDF Martijn Bastiaan, Christiaan Baaij (QBayLogic), Martin Izzard (Google DeepMind), Felix Klein (QBayLogic), Sanjay Lall (Google DeepMind, Stanford University) and Tammo Spalink (Stanford University)
Deadlock-free routing for Full-mesh networks without using Virtual Channels Download PDF Alejandro Cano Cos, Cristóbal Camarero, Carmen Martinez (Universidad de Cantabria) and Ramon Beivide (Universidad de Cantabria, Barcelona Supercomputing Center)
12:00 - 12:30 Lunch
12:30 Technical Paper Session B: Co-packaged Optics Session chair: Hossein Hashemi (University of Southern California, Astera Labs)
Accelerating Frontier MoE Training with 3D Integrated Optics Download PDF Taylor Groves, Mikhail Bernadskiy, Ho John Lee, Peter Carson, Thomas Graham and Eric Yeh (Lightmatter)
Silicon Photonic Accelerated Memory Pooling for Efficient Compute Resource Allocation Download PDF Zhenguo Wu and Keren Bergman (Columbia University)
13:30-13:45 Diamond Sponsor Talk: Cisco
Managing the fabrics underpinning AI infrastructure [details]
Will Eatherton (Cisco)
13:45-14:00 Diamond Sponsor Talk: Nubis
Overcoming Operational Barriers to AI Scale-Up with Co-Packaged Optics Integration [details] Download PDF
Peter Winzer (Nubis)
14:00 - 15:30 Panel – LLM Token Economy: How is networking going to play in the age of agents? [details]
Session chair: Alan Benjamin (GigaIO)
Moderator: Peter Olcott (First Spark Ventures)
Panelists: Phil Brown (Meta), Davor Capalija (Tenstorrent Inc.), Ethan Lockshin (Groq), Sid Sheth (d-Matrix)
15:30 The Head Bubba Memorial & HotI Closing Remarks Day 1 Dan Pitt (Palo Alto Innovation Advisors)

Conference Day 2: Thursday, August 21

Time Title Presenters
08:50 Day 2 Welcome General Chair: Artem Polyakov (NVIDIA)
09:00 - 10:00 Keynote 2: Evolving inter-connect fabric and SDN in the cloud to meet the needs of AI
[details]

Session chair: Dan Pitt (Palo Alto Innovation Advisors)
Deepak Bansal (Microsoft)
10:00 - 10:15 Diamond Sponsor Talk: Avicena
Paradigm Shift in AI Scale Up Clusters using microLED based Interconnects [details] Download PDF
Chris Pfistner (Avicena)
10:15 - 10:25 Platinum Sponsor Talk: Arista
Multiple Pathways to Energy Efficient Optics [details]
Andreas Bechtolsheim (Arista Networks)
10:25 - 10:30 Gold Sponsor Talk: GigaIO
GigaIO: Empowering Every Accelerator to Lead the AI Inference Revolution [details]
Alan Benjamin (GigaIO)
10:30 - 12:00 Technical Paper Session C:
Interconnect Software – Runtimes & Workloads
Session chair: Galen Shipman (Los Alamos National Laboratory)
Demystifying NCCL: An In-depth Analysis of GPU Communication Protocols and Algorithms Download PDF Zhiyi Hu, Siyuan Shen, Tommaso Bonato (ETH Zurich), Sylvain Jeaugey (NVIDIA), Cedell Alexander, Eric Spada (Broadcom), James Dinan, Jeff Hammond (NVIDIA) and Torsten Hoefler (ETH Zurich)
Characterizing Communication Patterns in Distributed LLM Inference Download PDF Lang Xu, Kaushik Kandadi Suresh, Quentin Anthony, Nawras Alnaasan and Dhabaleswar K Panda (The Ohio State University)
FPsPIN: An FPGA-based Open-Hardware Research Platform for Processing in the Network Download PDF Timo Schneider, Pengcheng Xu and Torsten Hoefler (ETH Zurich)
12:00 - 12:30 Lunch
Invited Talk Session: Next Gen Interconnects Session chair: Taylor Groves (Lightmatter)
12:30 - 13:00 Ultra Ethernet for next-generation AI and HPC workloads
[details]
Torsten Hoefler (ETH Zurich)
13:00 - 13:30 Building Custom AI Infrastructure with NVLink Fusion
[details]
Krishnan Geeyarpuram (NVIDIA)
13:30 - 13:40 Platinum Sponsor Talk: UALink
Why UALink: A Protocol-Level Comparison for Rack-Scale AI [details]
Sharada Yeluri (Astera Labs)
13:40 - 13:50 Platinum Sponsor Talk: Celestial AI
Photonic Fabric: Optical Scale-Up Networks for Accelerated Computing [details]
Ravi Mahatme (Celestial AI)
13:50 - 14:00 Platinum Sponsor Talk: Meta
AI Infra at Meta [details]
Wesley Bland (Meta)
14:00 - 15:00 Technical Paper Session D: Scale-up Interconnects Session chair: Letizia Giuliano (Alphawave Semi)
Micro-LED data interconnect for scale-up networks with record energy efficiency Bardia Pezeshki, Alex Tselikov, Rob Kalman, Suresh Rangarajan, Ivan Huang, Jeff Pepper, Sarah Zou, Howard Rourke and Alasdair Fikouras (AvicenaTech Corp)
On-Package Memory with Universal Chiplet Interconnect Express (UCIe): A Low Power, High Bandwidth, Low Latency and Low Cost Approach Download PDF Debendra Das Sharma, Swadesh Choudhary, Peter Onufryk (Intel) and Robert Pelt (AMD)
15:00 HotI Awards Yiltan Temucin (AMD) and Ryan Grant (Queen’s University)
15:10 HotI Closing Remarks Vice Chair: Sayan Ghosh (Pacific Northwest National Laboratory)

Tutorial:
Conference Day 3: Friday, August 22

Track 1

Time Title Presenters
Morning Sessions Session chair: Mike Adams (University of New Mexico)
08:30 - 12:00 Principles and Practice of Scalable and Distributed Deep Neural Networks Training and Inference
[details]
Dhabaleswar K. (DK) Panda, Nawras Alnaasan (The Ohio State University)
12:00 - 13:00 Lunch
Afternoon Session Session chair: Nicole Avans (Tennessee Technological University)
13:00 - 16:30 High-Performance and Smart Networking Technologies for HPC and AI
[details]
Dhabaleswar K. (DK) Panda, Benjamin Michalowicz (The Ohio State University)

Track 2

Time Title Presenters
Morning Sessions A Session chair: Matthew Dosanjh (Sandia National Laboratories)
7:00 - 8:30 Libfabric Communication Library
[details]
Rajalaxmi Angadi, Alexia Ingerson (Intel)
Morning Sessions B Session chair: Jackson Wesley (UNM)
08:30 - 12:00 GPU Communication Libraries for Accelerating HPC and AI Applications
[details]
Benjamin Glick Pouya Kousha, Arnav Goel (NVIDIA), Andreas Herten (Jülich Supercomputing Centre)