Hot Interconnects, San Jose, CA (2026)

08/10/2026

🚀 The IEEE Hot Interconnects 2026 Technical Program is Now Live! (Register: https://lnkd.in/dsVsUr8B)

We are excited to announce that the IEEE Hot Interconnects 2026 technical program is now available!

This year, HotI brings together leading researchers and industry experts to discuss the latest advances in high-performance interconnects, AI infrastructure, optical networking, memory systems, data center architectures, and next-generation networking technologies.

The program features inspiring keynotes, an engaging panel discussion, cutting-edge technical paper sessions, our new Lightning Talks session highlighting emerging ideas, and highly technical sponsor talks showcasing the latest industry innovations.

📅 Explore the full program here: https://hoti.org/2026/program.html

We look forward to welcoming you to IEEE Hot Interconnects 2026!

07/29/2026

𝐒𝐩𝐨𝐧𝐬𝐨𝐫 𝐓𝐚𝐥𝐤: Ravi Mahatme (Marvell Technology) "A Shared Memory Architecture for Token-Efficient AI Infrastructure" (register: https://hoti.org/2026/attendee_instructions.html)

We are excited to feature a sponsor talk from Ravi Mahatme, Senior Director, Product Management, Photonic Fabric Business Unit at Marvell Technology.

𝐀𝐛𝐬𝐭𝐫𝐚𝐜𝐭:
As AI inference scales, infrastructure performance is increasingly constrained by the ability to move data efficiently between compute and memory rather than by compute alone. Traditional server architectures, built around fixed memory-to-compute ratios, limit memory capacity, infrastructure utilization, and overall AI efficiency.

This presentation introduces a shared memory architecture that enables memory to scale independently of compute while maintaining the low latency and high bandwidth required for modern AI workloads. It examines the architectural principles behind a new shared memory tier, the role of optical connectivity in extending memory across servers and racks, and the system-level considerations required to deliver near-local memory performance at pod scale.

Attendees will gain insight into how this architectural approach improves infrastructure utilization, reduces unnecessary data movement and recomputation, and enables more token-efficient AI systems. The session also explores how advances in optical interconnect technology are making shared memory practical for next-generation AI infrastructure.

𝐁𝐢𝐨𝐠𝐫𝐚𝐩𝐡𝐲:
Ravi Mahatme is a product and systems leader focused on building next-generation AI infrastructure at the intersection of silicon, photonics, systems, and software. He joined Marvell Technology through its acquisition of Celestial AI and is currently Senior Director of Product, where he leads product strategy for rack-scale AI memory and interconnect platforms based on Photonic Fabric™ optical technology. Prior to Celestial AI, Ravi worked on distributed training and profiling capabilities for large-scale GPU workloads at AWS SageMaker. At Arm, he led product management for Ethos NPUs and spent more than a decade across engineering, business development, and product leadership roles, helping bring processor, accelerator, and platform technologies from concept to deployment. Ravi holds an M.S. in Computer Engineering from North Carolina State University and a B.E. in Electronics Engineering from the University of Mumbai.

Don’t miss this opportunity to learn how shared memory architectures and optical interconnects are enabling more token-efficient AI infrastructure. Join us at IEEE Hot Interconnects 2026 to explore the technologies shaping the future of AI networking.

Follow us on Facebook for updates on IEEE Hot Interconnects.

IEEE Computer Society

07/24/2026

🚀𝐒𝐩𝐨𝐧𝐬𝐨𝐫 𝐓𝐚𝐥𝐤: Alex Omø Agerholm (Napatech) "AI Fabrics in Motion: Programmable Interconnects for Scale-Up, Scale-Out, and Scale-Across" (register: https://hoti.org/2026/attendee_instructions.html)

We are excited to feature a sponsor talk from Alex Omø Agerholm, Chief Product Architect at Napatech.

𝐀𝐛𝐬𝐭𝐫𝐚𝐜𝐭:
AI moved the bottleneck from compute to the interconnect. As fabrics scale up, out, and across, they meet traffic they were never built for — while AI evolves faster than silicon can be re-spun. This talk makes the case for programmable infrastructure: networking that adapts at software velocity.

𝐁𝐢𝐨𝐠𝐫𝐚𝐩𝐡𝐲:
Alex Omø Agerholm is Chief Product Architect at Napatech. He is a highly experienced product and system architect with an entrepreneurial mindset and a "yes we can" attitude. Known for his technical depth, clear communication across all levels, and a relentless drive to improve teams, products, and processes, Alex combines deep technical expertise with a practical approach to product and system architecture.

Don’t miss this opportunity to hear how programmable interconnects are enabling the next generation of AI infrastructure. Join us at IEEE Hot Interconnects 2026 to explore the technologies shaping the future of AI networking.

Follow us on Facebook for updates on IEEE Hot Interconnects.

IEEE Computer Society

07/21/2026

𝐊𝐞𝐲𝐧𝐨𝐭𝐞: Omar Baldonado (Meta) "Lessons from Networking Meta’s Gigawatt-Scale AI Fleet" (register: https://hoti.org/2026/attendee_instructions.html)

We are excited to feature a keynote presentation from Omar Baldonado, Senior Director of Networking at Meta.

𝐀𝐛𝐬𝐭𝐫𝐚𝐜𝐭:
Networking for AI Infrastructure has never evolved so quickly as it is now. This is especially true for Meta, who drives a gigawatt-scale fleet of AI accelerators across a wide range of model use cases and product groups; accelerator chips and systems; and networks and network hardware.

In this talk, Omar Baldonado, Senior Director of Networking at Meta, will reveal how Meta’s network team meets the aggressive time-to-market needs of AI capacity and the lessons learned from introducing new products and technologies into the fleet every quarter. Specifically, at Meta’s gigawatt scale, we treat this as a cross-network-domain problem, driven by model use cases and accelerator capabilities, that places requirements on the overall network. The team is constantly and quickly co-designing across multiple layers of the stack and across the whole AI infrastructure to (1) optimize performance, (2) ease technology and capacity transitions, and (3) provide flexibility. Because we are always introducing new accelerator hardware types, new network hardware and topologies, and new locations into the fleet, we consider all three of these goals simultaneously. While scale-up, scale-out, and scale-across networks utilize different technologies, they must be treated integrally in order to address these goals.

The talk is grounded in Meta’s experience designing, building, and operating the scale-up, scale-out, and scale-across networks for Prometheus, Meta’s first gigawatt-scale cluster, over the last 1-2 years. Omar will also cover learnings from working with both hardware and cloud vendors to leverage open standards to move faster in providing AI capacity to the rest of Meta.

𝐁𝐢𝐨𝐠𝐫𝐚𝐩𝐡𝐲:
Omar Baldonado leads the groups that develop and operate Meta's global data center networks. These networks support all of Meta’s AI models and the Meta family of apps (AI at Meta, Facebook, Instagram, WhatsApp, Messenger). His teams have developed some of the largest AI clusters in the world, with gigawatt-scale clusters on the way, while contributing to open-source projects including TorchComms for PyTorch and FBOSS for switches, and collaborating through the Open Compute Project. With over three decades of experience in networking, Omar will share insights into the networking architectures and operational lessons behind Meta’s gigawatt-scale AI infrastructure.

Don’t miss this opportunity to hear directly from one of the leaders building AI networking infrastructure at unprecedented scale. Join us at IEEE Hot Interconnects 2026 to shape the future of AI infrastructure.

Follow us on Facebook for updates on IEEE Hot Interconnects.

IEEE Computer Society

07/06/2026

🎤Keynote: Gilad Shainer (NVIDIA) "Networking Innovations for Gigascale AI Systems" (register: https://hoti.org/2026/attendee_instructions.html)

We are excited to feature a keynote presentation from Gilad Shainer, Senior Vice President of Networking at NVIDIA.

Abstract: As AI factories grow to hundreds of thousands, and soon millions of interconnected GPUs, the network has become the defining architecture of large-scale AI. Achieving uncompromised performance requires fabrics that scale up, scale-out and scale-across. In this session, we will discuss the full breadth of networking innovations that enable the construction of gigascale AI infrastructure, delivering the performance required to train massive foundation models, deploy real-time inference pipelines, and interconnect federated AI factories globally.

With more than two decades of leadership in high-speed networking and AI infrastructure, Gilad has played a pivotal role in advancing technologies that power today's largest AI systems. A recipient of multiple R&D 100 Awards for innovations including CORE-Direct In-Network Computing and UCX, Gilad will share insights into the networking breakthroughs driving the next generation of gigascale AI infrastructure.

Don’t miss this opportunity to hear directly from one of the industry's leading innovators in AI networking and large-scale computing. Join us at IEEE Hot Interconnects 2026 and be part of the conversation shaping the future of AI infrastructure.

Follow us on Facebook for the latest updates from IEEE Hot Interconnects.

IEEE Computer Society

06/24/2026

🎤 Sponsor talk: Will Eatherton (Cisco) “Ethernet for Shared AI Fabrics: InfiniBand-Class Performance with Multi-Tenant Scale” (register: https://hoti.org/2026/attendee_instructions.html)

Will Eatherton, SVP, Data Center, Internet & Cloud Infrastructure Engineering at Cisco, will deliver this Sponsor Talk at IEEE Hot Interconnects 2026. The talk will focus on one of the key questions shaping modern AI and HPC infrastructure: how can shared, open, multi-tenant Ethernet fabrics deliver the performance, isolation, and scalability required by large-scale distributed training workloads?

The session will discuss the architectural choices and enabling technologies that are pushing Ethernet toward InfiniBand-class performance, including RoCEv2, SRv6, Priority Flow Control, adaptive routing, packet-level multipathing, hardware-assisted transport, and programmable congestion control. It will also highlight how tenant isolation, integrated multi-job scheduling, congestion avoidance, and intelligent traffic management are becoming essential for production AI clusters.

As AI infrastructure continues to move from isolated proprietary deployments toward shared open fabrics, this discussion will offer a timely perspective on the role of Ethernet, Ultra Ethernet, and open standardization in next-generation accelerator-to-accelerator communication.

📅 IEEE Hot Interconnects 2026
🗓️ 19–21 August 2026

IEEE Computer Society

06/18/2026

🎤 Keynote: Bilal Riaz (Ciena) "The Future of AI Interconnects: Open Approaches to High-Performance AI Infrastructure" ( register: https://hoti.org/2026/attendee_instructions.html )

We are excited to feature a keynote presentation from Bilal Riaz, Senior Director of Product Line Management and Head of Interconnect Strategy at Ciena.

Abstract: As AI training and inference workloads continue to expand in both scale and complexity, the traditional definitions of scale-up, scale-out, and scale-across architectures are increasingly blurring. Modern “scale-across” approaches distribute high-capacity computation across multiple sites, intensifying familiar interconnect challenges such as jitter, burstiness, congestion, and long-tail latency. At the same time, the historical segmentation between IMDD and coherent optics is blending—mirroring the way electrical-versus-optical boundaries are disappearing within the rack itself. As AI workloads push unprecedented compute density and interconnect bandwidth, existing scale-up, scale-out, and scale-across architectures are all evolving to meet new performance and connectivity demands. This talk will explore how technologies across these various scaling domains are converging to meet the evolving needs of next-generation customer applications.

With extensive experience leading the development of next-generation optical modem technologies and coherent networking innovations, Bilal brings a unique perspective on the future of AI connectivity and infrastructure. His insights will highlight how open high-performance interconnect technologies are enabling the next wave of scalable AI systems.

Don’t miss this opportunity to hear directly from a leader at the forefront of optical networking and AI infrastructure innovation. Join us at IEEE Hot Interconnects 2026 and be part of the conversation shaping the future of AI interconnects.

Follow us on Facebook for the latest updates from IEEE Hot Interconnects.

IEEE Computer Society

06/10/2026

📢 Deadline Extended: IEEE Hot Interconnects (HotI) 2026 Call for Tutorials — Proposals Now Due June 12, 2026!

There's still time to submit your tutorial proposal for IEEE Hot Interconnects (HotI) 2026!

HotI 2026 will take place virtually from August 19–21, 2026, with tutorials held on August 21, 2026.

We invite proposals for half-day and full-day tutorials exploring the technologies, systems, and design choices shaping modern interconnects across long, short, and ultra-short distances.

Topics may include:
🔹 AI/ML networking
🔹 Data center interconnects
🔹 Optical fabrics
🔹 Chiplet and accelerator interconnects
🔹 Congestion management
🔹 SDN and overlays
🔹 Traffic characterization
🔹 Fault tolerance
🔹 High-bandwidth, low-latency I/O

Hands-on segments are strongly encouraged to make tutorials more engaging and practical.

🗓 Updated Key Dates
• Proposal Deadline (Extended): June 12, 2026
• Notification of Acceptance: June 19, 2026
• Materials Due: July 31, 2026

Submissions are handled through EasyChair: https://easychair.org/account/signin?l=4202408119257879241.1781125246.1f8a68e1

We look forward to receiving your tutorial proposals and seeing you at HotI 2026! 🚀

👉 Full Call for Tutorials: https://hoti.org/2026/call-for-tutorials.html
📩 Questions: [email protected]

IEEE Computer Society

06/04/2026

🚀 Registration is now open for IEEE Hot Interconnects 2026!

📅 August 19–21, 2026
🌐 Virtual via Zoom Events
🎟️ Free attendance

You can find the registration instructions here:
https://hoti.org/2026/attendee_instructions.html

After registering, attendees will receive their personalized Zoom Events ticket and joining information by email.

Questions about registration? Contact us at [email protected]

We look forward to seeing you at HotI 2026!

05/27/2026

🚀 We are excited to introduce the Technical Program Committee for IEEE Hot Interconnects 2026.

This year’s TPC brings together leading experts from academia, national laboratories, and industry, spanning high-performance interconnects, datacenter networking, AI infrastructure, memory systems, accelerators, optical and electrical fabrics, and large-scale distributed systems.

Their expertise will be essential in shaping a strong technical program and selecting the papers that capture the latest advances and emerging directions in the field.

A big thank you to all TPC members for their time, service, and contribution to the HotI community! 🙌

🔹 2026 Technical Program Committee
Khaled Diab, Hewlett Packard Enterprise
Kapil Shrikhande, Upscale AI
Weiyang Wang, Massachusetts Institute of Technology
Yuyang Wang, University of Connecticut
Scott Hemmert, Sandia National Laboratories
Charles Clayton, NVIDIA
Xi Wang, University of California, Merced
Jongryool Kim, SK hynix Inc.
Yiwei Yang, University of California, Santa Cruz
Aurelien Bouteiller, AMD
Luke Chang, AMD
Peter Del Vecchio, Self
Shinji Sumimoto, The University of Tokyo
Ferrol Aderholdt, NVIDIA
Yuichiro Ajima, Fujitsu Limited
Ron Brightwell, Sandia National Laboratories
Swadesh Choudhary, Intel
Hans Eberle, NVIDIA
Pedro Javier Garcia, Universidad de Castilla-La Mancha
Hyun-Wook Jin, Konkuk University
Jithin Jose, Microsoft
Shelby Lockhart, AMD
Xiaoyi Lu, University of Florida
Mohiuddin Mazumder, Intel Corporation
Md Atiqul Mollah, Cornelis Networks
Kishore Punniyamurthy, NVIDIA
Kaushik Kandadi Suresh, NVIDIA
Raghunath Raja Chandrasekar, Amazon Web Services
Adrien Roussel, CEA
Craig Stunkel, NVIDIA
Hugo Taboada, CEA
Madeleine Glick, Columbia University
KC Liu, AWS

Thank you to the full committee for helping make Hot Interconnects 2026 possible. 🔥

Hot Interconnects

08/10/2026

07/29/2026

07/24/2026

07/21/2026

07/06/2026

06/24/2026

06/18/2026

06/10/2026

06/04/2026

05/27/2026

Address

Telephone

Website

Alerts

Shortcuts

Share

Category

Listing takedown request