S16 【S16】Next-Generation AI Server & Advanced Liquid Cooling (TTMA)
Oct. 20, 2026 15:50 PM - 17:50 PM
Room: 701A, 7F, TaiNEX 2
Session chair： Edward Kung, TTMA / Jimmy Chuang, Intel
From Chip to AI Factory: A Five-Year Roadmap for Silicon Suppliers and Hyperscalers
發表編號：S16-1時間：15:50 - 16:20 |
Invited Speaker
Speaker: Gamal Refai-Ahmed, Senior Fellow and Chief Thermo-Mechanical Architect, AMD
Bio:
Gamal Refai-Ahmed, Ph.D., FIEEE, LFASEM, is a distinguished technical executive and member of the National Academy of Engineering with over 20 years of experience at AMD, GE, and Cisco. His expertise includes thermal management, power architecture, and advanced packaging. He has authored over 120 publications and holds more than 160 granted or pending patents.
Abstract:
AI infrastructure is evolving from accelerator-level optimization toward scalable AIfactory deployment. Peak silicon performance alone is no longer sufficient; deployment depends on the integration of silicon, advanced packaging, substrates, boards, racks, liquid cooling, power delivery, software, and regional infrastructure. This talk presents a five-year roadmap centered on the validated envelope—a multidomain operating range in which chip-to-facility systems can be manufactured, powered, cooled, monitored, serviced, and replicated at scale. It will address key technology drivers, including chiplets, HBM, optical I/O, power delivery, liquid cooling, mechanical reliability, and fleet telemetry, while proposing validation milestones for silicon suppliers, hyperscalers, foundries, packaging ecosystems, and infrastructure providers. The goal is to enable a “validate once, replicate many times” approach to scalable AI infrastructure.
|
|
AI Industry Trending and Challenges
發表編號：S16-2時間：16:20 - 16:50 |
Invited Speaker
Speaker: Gerry Juan, Senior Systems and Hardware Enablement Architect, Intel
Bio:
Gerry Juan is a Senior Systems and Hardware Enablement Architect at Intel, specializing in AI infrastructure, data center platforms, and high-performance computing systems. He is part of Intel's Datacenter Customer Engagement organization, where he collaborates closely with ODM/OEM partners and cloud customers to enable next-generation AI and hyperscale computing platforms.
Abstract:
This presentation discusses the thermal and energy efficiency challenges of next-generation AI infrastructure, including rack-scale AI systems, heterogeneous CPU-GPU architectures, memory hierarchy optimization, and cooling considerations for Agentic AI workloads. The session will explore system-level approaches to improve performance per watt while addressing increasing power density in modern AI datacenters.
|
|
From Chip to Rack: Evolution of Thermal Management Architecture
發表編號：S16-3時間：16:50 - 17:20 |
Invited Speaker
Speaker: Rechan Fan, Sr. Engineer, Lenovo
Bio:
Senior Thermal Engineer at Lenovo with over 25 years of experience in server thermal engineering, specializing in high-power CPU/GPU thermal solutions, rack-level cooling architecture, and data center cooling solutions.
Abstract:
Rapidly increasing power density is changing the scale of thermal management—from high-power chips to high-density racks. This talk explores how thermal, flow, and energy considerations must be co-optimized as liquid cooling evolves from a component-level heat removal technique to a system-level thermal architecture.
|
|
Liquid Cooling: How to Achieve 100% Heat Capture on 500 kW Racks
發表編號：S16-4時間：17:20 - 17:50 |
Invited Speaker
Speaker: James Wu, Team Lead, Thermal Mechanical Engineer, CoolIT, An Ecolab Company
Bio:
James Wu is the Thermal Mechanical Engineering Team Lead from CoolIT, focusing on direct liquid cooling (DLC) solutions for high-power AI/HPC platforms. He is passionate about pushing the boundaries of server cooling solutions for coldplate, DIMM channel, conduction plate, and full-system liquid cooling designs. He will share the insights and experience on achieving 100% heat capture based on CoolIT's design philosophy and real-world practices. He is open to discuss liquid cooling innovations, structural design and data center thermal management with fellow thermal/mechanical engineers and system architects.
Abstract:
As rack power densities surge beyond 100 kilowatts (kW), achieving 100% heat capture becomes increasingly essential for thermal management and operational efficiency. Current coldplate designs focus on high-thermal design power components such as CPUs and GPUs, but future architectures demand complete heat removal across all components. Even at 90% heat capture, a 500kW rack leaves 50kW for air cooling, which is a stretch for even the best air-cooled implementations. Enabling 100% heat capture will be critical for supporting next-generation AI infrastructure. Join us to learn how integrated cooling loops and advanced coldplate technologies enable complete heat removal across the entire server.
|
|
|