Vlad Rozanovich, Senior Vice President of Lenovo Infrastructure Solutions Group, shares his insights from Tech World 2024
- Delivering AI for all requires setting a new bar for sustainable operations: 6th generation Lenovo Neptune liquid cooling supercharges efficiency to bring NVIDIA Blackwell platform and trillion parameter AI to any enterprise
This week, I had the privilege of taking part in unveiling the next phase of our Smarter AI for All vision on behalf of the Infrastructure Solutions Group at Lenovo’s annual global innovation event, Lenovo Tech World ’24 , in Seattle. For those of us in attendance, it was an incredible firsthand look at the comprehensive solutions, services, and devices that are accelerating AI and delivering transformation for customers and industries throughout the world.
Joined by an unparalleled community of partners and customers, Lenovo showed how AI has definitively moved beyond potential and into transformative results. Our combination of expertise, infrastructure, and an ever-evolving, interconnected ecosystem was on full display with live demonstrations of how our technology is helping businesses everywhere reimagine workflows, unleash creativity, and turn unstructured data into actionable insights.
With our extensive global ecosystem and network of partners, like NVIDIA, we are designing the next generation of data centers and AI solutions that support the surge in data while more efficiently powering AI compute and unlocking its potential to support more sustainable outcomes.
Supercharging Efficiency to Power Surging Demand
As AI adoption increases, data centers need to be re-engineered for higher thermal densities. Lenovo is continuing our unwavering commitment to secure, efficient hybrid infrastructure at every point, from edge to cloud, and to setting the bar for sustainable high-performance computing (HPC) operations. To that end, we unveiled game-changing technology innovations and joint solutions that usher in a new era of power efficiency and accessibility for AI-level compute.
For more than a decade, Lenovo Neptune has empowered businesses to harness direct liquid cooling to achieve up to 40 percent lower power consumption and higher density computing. Today, we announced the 6th generation of Neptune liquid cooling systems that are reshaping water cooling and pushing industry boundaries even further with the breakthrough ThinkSystem N1380. This engineering innovation puts the highest accelerated computing available in a compact chassis any size organization can use.
We have reimagined and engineered the rack for vertical computing nodes to optimize space and increase total heat removal. The infrastructure fits within less than a square meter of data center space – in industry-standard 19″ racks, pushing the boundaries of compact general-purpose computing. With Neptune, customers can direct 30% more power for compute, maximizing it for AI and research rather than cooling.
It uses Neptune’s warm water system. Unlike many others in the market, Neptune doesn’t have a chiller before the rack, wasting electricity. 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.
In his keynote, our Lenovo Chairman and CEO Yuanqing Yang was joined by NVIDIA founder and CEO Jensen Huang to unveil the new Lenovo ThinkSystem SC777 V4 Neptune, a powerful new system that supercharges efficiency using the 6th generation Neptune innovation and brings the NVIDIA Blackwell platform to enterprises of any size.
Marking a leap forward in sustainability and energy efficiency for AI systems, this 100% liquid-cooled system requires no fans or specialized data center air conditioning. It fits into a standard rack and runs on standard power. What’s more, this innovation is so pivotal, that it can be used to efficiently power AI at every scale.
Yang and Huang emphasized the critical need for energy-efficient AI infrastructure.
“Speed is sustainability. Speed is performance. Speed is energy efficiency,” Huang said, stressing how performance improvements directly contribute to reducing energy consumption and increasing efficiency.
“Lenovo’s 6th Generation Neptune Liquid Cooling solution supports AI computing and high-performance computing while delivering better energy efficiency,” Yang said.
Now rolling out across our partner ecosystem, the new system underscores how Lenovo is pioneering data centers in which 100KW+ server racks are running without the need for specialized data center air conditioning. As part of an expanded collaboration to deliver full-stack hybrid AI that drive speed, innovation and productivity across the enterprise the new liquid cooling design enables organizations of any size to build and leverage AI-compute one tray at a time, using standard power and an open ecosystem built on the world’s most widely used cooling system.
With these advancements Lenovo is pioneering a new age of power efficiency, allowing organizations of all sizes to harness AI without the burden of specialized data center cooling requirements. Through a collaborative ecosystem, Lenovo is reshaping the landscape for sustainable AI computing, ensuring that as AI adoption grows, it does so with the intent of minimizing environmental impact.
You can read more about this incredible story on Lenovo StoryHub or visit Smarter AI for All | Tech World ’24 | Lenovo US to learn more about the innovations from Tech World.