Site icon AIT365

ZutaCore Unveils the First Waterless, Direct-to-Chip Liquid Cooling Cold Plates for NVIDIA’s GB200 Grace Blackwell Superchip at Dell Technologies World 2024

ZutaCore

With A Single, Monolithic Cold Plate Capable of Cooling 2,800 Watts and Beyond, ZutaCore’s HyperCool Technology Ushers in a New Era of Sustainable AI for the Masses

ZutaCore, a leading provider of direct-to-chip, waterless liquid cooling solutions, announced support for the NVIDIA GB200 Grace Blackwell Superchip to enable hyperscalers to harness the massive processing power of the world’s most powerful AI GPU both sustainably and cost-effectively. At next week’s Dell Technologies World, ZutaCore will be showcasing the industry’s first dielectric cold plates to support this new platform, featuring an unprecedented ability to cool 120kW of rack power with little or no modifications to current real estate, power, or cooling systems.

With over 400 billion transistors, the Grace Blackwell superchip features two NVIDIA Blackwell GPUs, an NVIDIA Grace CPU, and two I/O chips. A single ZutaCore HyperCool® dielectric cold plate will sit directly on top of this superchip, taking up very little space while providing significant cooling capabilities. In a typical GB200 2RU (featuring two 2 x NVIDIA GB200 modules of 2,800 watts each) a single, monolithic ZutaCore cold plate can deliver 2,800 watts of cooling. The monolithic cold plate ensures a simple design with only one input and one output, which reduces the potential points of failure to deliver a higher level of reliability. The module is also equipped with OCP-compliant blind mate connectors that support 5,600kW.

“ZutaCore’s groundbreaking technology enables us to cool the NVIDIA GB200 Superchip with a single cold plate, delivering an industry-first waterless direct-to-chip liquid cooling solution,” said Erez Freibach, Co-founder and CEO at ZutaCore. “This breakthrough frees AI data center owners and operators from the risks associated with water-based cooling, empowering them to deploy the processing power needed for compute-intensive workloads like generative AI without the fear of leakages.”

Also Read: Avant Technologies Equipping AI-Managed Data Center with High Performance Computing Systems

“The next generation of high-performance GPUs, such as NVIDIA’s GB200, that are required to meet the demanding needs of AI model training will generate a significant amount of heat, making efficient cooling solutions crucial for optimal system performance and sustainable operation,” said Mark Nossokoff, Research Director for Hyperion Research. “Innovative cooling solutions such as ZutaCore’s HyperCool waterless, direct-to-chip liquid cooling approach that eliminates the need for water in the cooling process, aim to optimize performance, maximize energy efficiency, and reduce operation costs associated with running the most advanced AI computing systems.”

The Power of HyperCool

Featuring a groundbreaking closed-loop system that operates at low pressure and moves large amounts of heat off the processors and away from the servers, ZutaCore’s HyperCool can be implemented in new or existing data centers to deliver 10 times more computing power, a 50% reduction in total cost of ownership, 100% heat reuse, and reduced CO2 emissions for a sustainable data. There is also a growing ecosystem of servers certified to work with HyperCool, including industry powerhouses such as Dell Technologies, ASUS, Pegatron, and SuperMicro.

The Growing NVIDIA Ecosystem Around HyperCool

In addition to HyperCooling the GB200 Grace Blackwell Superchip, ZutaCore previously unveiled support for the NVIDIA H100 and H200 GPUs. Further strengthening this ecosystem, ZutaCore announced in a separate press release issued today that it has secured a strategic OEM agreement with UNICOM Engineering (a Dell Titanium Partner) to deliver warrantied HyperCooled AI servers for the H100 and H200 GPUs. ZutaCore also announced that a global AI-as-a-Service (AIaaS) leading provider has already chosen its HyperCool direct-to-chip liquid cooling for AI servers with deployments starting in Q3 2024. The first platform will be the Dell Technologies XE9680, which will be delivered by leveraging UNICOM Engineering’s global services.

Source: PRNewsWire

Exit mobile version