Lenovo Tech World 2024: Liquid cooling gets cooler

Lenovo unveiled a new 6th generation vertical liquid cooling system, the ThinkSystem N1380 Neptune, at the recent Lenovo Tech World 2024 innovation event in Seattle.

Designed to support accelerated computing for generative AI, the system reduces data centre power consumption by up to 40%.

Lenovo also expanded its longstanding collaboration with Nvidia by announcing Lenovo Hybrid AI Advantage with NVIDIA and unveiled the ThinkSystem SC777 V4 Neptune on stage. This new system leverages the new ThinkSystem N1380 Neptune to help bring Nvidia Blackwell and the Nvidia GB200 to the enterprise market, enabling trillion-parameter AI models in a compact design. 

As AI adoption increases, data centres need to be re-engineered for higher thermal densities. Lenovo’s ThinkSystem N1380 Neptune and ThinkSystem SC777 V4 Neptune enable data centres to run 100KW+ server racks without specialised air conditioning. This development supports denser server setups and allows for power repurposing to cooling, aligning with the demands of AI-driven data centres.

“For more than a decade, Lenovo has pioneered liquid cooling innovations, with the goal of bringing the power of high-performance computing to every organization,” said Yuanqing Yang, Lenovo chairman and CEO.

“Through groundbreaking liquid-cooling engineering and technology integration, ThinkSystem N1380 Neptune is unlocking a new era of data centre efficiency and design that will help bring Nvidia Blackwell and trillion-parameter AI to all, while fundamentally changing how power is used in the data centre.”

With ThinkSystem N1380 Neptune, data centre power is increased for AI and research, not for system cooling. The technology uses a reimagined vertical liquid cooling chassis to deliver accelerated computing with advanced efficiency in a compact system that is 100% liquid cooled, eliminating the need for power-consuming fans. Designed for industry standard 19-inch racks, organisations can leverage the highest accelerated computing available using standard power in an open ecosystem. 

Bob Pette, VP of enterprise platforms at NVIDIA, said: “The Nvidia Blackwell platform is the engine to power generative AI and define a new industrial revolution.

“With the breakthrough capabilities of Lenovo ThinkSystem N1380 Neptune and Nvidia GB200, we’re supporting the world’s most technologically advanced accelerated workloads while massively reducing operation costs and energy consumption to help realize the promise of AI for every industry.”

The ThinkSystem SC777 V4 Neptune supports the next-generation Nvidia Quantum-X800 InfiniBand  and Spectrum-X800 Ethernet platforms for high-performance accelerated networking. In addition, the ThinkSystem SC777 V4 Neptune supports Nvidia AI Enterprise, a cloud-native software platform that streamlines development and deployment of production-grade AI solutions, including generative AI, computer vision, and speech AI.

Bringing Nvidia GB200 to the enterprise market with 100% heat removal

Lenovo will deliver Nvidia GB200 rack systems that supercharge AI training, data processing, engineering design and simulation. Featuring open-loop, direct warm-water cooling for the entire Nvidia GB200 Grace Blackwell system architecture with NVIDIA NVLink interconnect, enterprise customers of any size can use the Lenovo ThinkSystem SC777 V4 Neptune to run trillion-parameter AI models.

With the advanced water cooling, the new ThinkSystem server allows critical components to operate at lower temperatures and effectively removes all heat from all components, including GPUs, memory, I/O, local storage, and voltage regulators. Combined, the innovations enable AI training and real-time LLM inference for models scaling up to 10-trillion parameters while increasing performance, energy-efficiency, and reliability in a compact, quiet system.

Scalable HPC and AI supercomputer

Lenovo helps organisations deploy high-performance AI at any scale. Scaling from a compact 13U enclosure to full racks, rows, and datacentres, the ThinkSystem N1380 Neptune features eight tray slots, four 15kW power conversion stations, with a sophisticated Neptune water flow distribution, packaged for optimal space, power, and cooling efficiency:

• Maximised standard form factor: Engineered from the ground up to provide the additional space needed without breaking data centre standards, N1380 maintains scalability to make the newest and highest performing technologies available to customers of every size.
• Reduced power consumption: The design eliminates the need for internal airflow and power-consuming fans, reducing power consumption by up to 40% compared with similar air-cooled systems in a typical data centre. Rack insulation significantly minimises radiant heat.
• Power conversion station (PCS): Beyond the Neptune water-cooling infrastructure, the N1380 enclosure houses up to four ThinkSystem 15kW Titanium Power Conversion Stations (PCS), supplying internal system power to a 48V busbar. This innovative design merges power conversion, rectification, and distribution into a single PCS, a departure from traditional setups that demand separate units, resulting in best-in-class efficiency.
• Reliable cooling: N1380 features an integrated manifold with a patented blind-mate mechanism and aerospace-grade dripless connectors to the compute trays for safer operation.
• Complete heat removal: With enhanced waterflow safety, the efficient connections facilitate 100% heat removal.
• Power repurposed: Neptune is designed to operate at water inlet temperatures as low as the dew point allows, up to 45°C, which eliminates the need for additional chilling and allows for efficient reuse of the generated heat for building heat or adsorption chilling cold water generation.

Over a decade of liquid cooling

Lenovo has been a leader in liquid cooling technology for more than a decade, removing heat and reducing power consumption in a different way compared to traditional air-cooled systems. With more than forty patents and the most widely used liquid cooling platform around the world, Lenovo has  finetuned its liquid cooling technology. Lenovo Neptune uses custom brazed copper water loops and patented CPU cold plates for full system water-cooling. Unlike systems that use low-quality FEP plastic, Neptune features durable stainless steel and reliable EPDM hoses.

Global network of Lenovo AI support

To realise the promise of AI for every industry, the 6^th generation Lenovo Neptune is available throughout the end-to-end hybrid AI portfolio of Lenovo servers, solutions, and services.

Businesses can build, scale and evolve private AI with the new GPU resources on demand, AI-driven systems management and advanced liquid cooling services that optimise Lenovo Neptune for one’s environment.

Lenovo TruScale GPUaaS, delivers fast GPU power on demand through a scalable, pay-as-you go model, while Lenovo Power and Cooling Services help organisations reduce the complexity of optimising demanding AI workloads. Lenovo TruScale GPUaaS will be available for a wide range of Nvidia Tensor Core GPU options, from NVIDIA H100 Tensor Core and L40S to the Nvidia GB200 with liquid cooling. This accelerates the availability of Nvidia accelerated computing to enable rapid real-world innovation, making AI use cases in the enterprise more accessible.