Lenovo and NVIDIA Partner on New Hybrid AI Solutions

Print Friendly, PDF & Email

[SPONSORED GUEST ARTICLE]  The recent NVIDIA GTC 2024 conference in San Jose was one of the most closely watched, news-filled and, generally speaking, electric technology conferences in recent memory. For three days, GTC was at the epicenter of the technology world with the focus, of course, on generative AI and its promise of generational impact on business, science, healthcare, entertainment and education, enabled by NVIDIA GPUs.

Lenovo is a critical partner combining NVIDIA chips within a range of data center server and workstation configurations designed to deliver transformational AI solutions that support organizations’ unique strategies.

The Lenovo booth, a sprawling affair at the center of the conference floor, was a center of interest at GTC. The company announced new hybrid AI solutions, built in collaboration with NVIDIA, powering tailored generative AI applications to enterprises and cloud.

“Lenovo solutions span from the pocket to the cloud,” said Kirk Skaugen, EVP and President of Lenovo’s Infrastructure Solutions Group, at the conference. “We go all the way to the largest data centers in the world, we truly have a multifaceted relationship with NVIDIA. In HPC we’re all over the TOP500 list of the world’s fastest supercomputers, including the most sustainable supercomputer on the GREEN500. So everything we’re doing with NVIDIA is important here at GTC.”

Lenovo unveiled the expansion of its ThinkSystem AI portfolio designed for the most demanding AI, data analytics and HPC workloads, featuring two powerful 8-way NVIDIA GPU systems combining massive computational capabilities with power efficiency. Engineered for generative AI, the Lenovo systems incorporate support for the NVIDIA HGX AI supercomputing platform, including NVIDIA H100 and H200 Tensor Core GPUs and the new Grace Blackwell GB200 Superchip announced at GTC, as well as the NVIDIA Quantum-X800 InfiniBand.

At GTC 2024

With impressive power efficiency (PUE) of 1.1., the new Lenovo ThinkSystem SR780a V3 is a 5U system that uses Lenovo Neptune™ liquid cooling. With direct water cooling of CPU and GPUs and NVIDIA NVLinkSwitch, the system can sustain maximum performance without hitting thermal limits. Neptune direct water-cooling uses warm water, delivering up to 40 percent reduction in power consumption and a 3.5x improvement in thermal efficiencies compared to air-cooled systems.  All that compute power is housed in a dense 5U package, helping to conserve valuable data center real estate.

The new Lenovo ThinkSystem SR680a V3 is an air-cooled, two-socket system built for complex AI with Intel processors and a choice of NVIDIA GPUs. It uses industry-standard 19-inch server racks, allowing for dense hardware configurations that maximize efficiency.

Lenovo also showcased a new Lenovo PG8A0N – the ultimate 1U node for AI and featuring open-loop liquid cooling for accelerators, supporting the NVIDIA GB200, which delivers 45X faster real-time LLM inference, 40X lower TCO and 40X less energy. Lenovo will deliver GB200 rack systems for AI training, data processing, engineering design and simulation.

Customers can run NVIDIA AI Enterprise, a cloud-native software platform for development of AI applications, on Lenovo’s NVIDIA-Certified Systems. For high-performance inference on AI models from NVIDIA and its ecosystem, customers can run NVIDIA NIM inference microservices.

The portfolio also supports the 4th and 5th generation of Intel Xeon Scalable Processors and provides thermal headroom for future higher-power GPUs.

Bringing NVIDIA Omniverse and HPC to CSPs and Enterprise

Lenovo is the leading provider of workstation-to-cloud support for the NVIDIA OVX systems and  NVIDIA Omniverse development platform. Lenovo is partnering with NVIDIA to build accelerated models faster using NVIDIA MGX modular reference designs. Leveraging the designs, cloud service providers receive customized models with the delivery of accelerated computing for AI and Omniverse workloads economically and at scale.

Lenovo’s new offerings built with the NVIDIA MGX modular reference server design include:

  • Lenovo HG630N – MGX 1U – an open-standard server with Lenovo Neptune direct liquid cooling.
  • Lenovo HG650N – MGX 2U – a modular, GPU-optimized, air-cooled system that supports industry-standard racks along with NVIDIA’s GH200 Grace Hopper Superchip.
  • Lenovo HG660X V3 – MGX 4U – a system that supports up to eight 600W NVIDIA GPUs in an air-cooled environment, ideal for NVIDIA Omniverse and AI workloads.
  • Lenovo HR650N – MGX 2U – an Arm CPU server with multiple cores and flexibility for storage and Front IO, leveraging the energy-efficient NVIDIA Grace CPU Superchip and supporting DPUs.

Lenovo ThinkSystem SR780a V3

Lenovo’s data science workstations offer up to 4x NVIDIA RTX 6000 Ada Generation GPUs for large AI training, inferencing and graphics-intensive workloads. This offers greater AI developer productivity through automated workflows:

  • New Lenovo Workstations with NVIDIA AI Workbench is a software tool that enables the development and deployment of powerful AI solutions for inferencing, large-scale simulation, and demanding industrial and scientific workflows. NVIDIA AI Workbench makes generative AI and machine learning development easier for developers.
  • Now available in Lenovo ThinkStation and ThinkPad workstations, the new NVIDIA A800 GPU designed specifically for AI enables secure, personal data science and Gen AI environments.

Lenovo ThinkStation PX with dual CPUs and up to 4x NVIDIA RTX professional graphics cards is the most powerful workstation on the market today.

Professional Services: Fast Tracking the AI Journey

Lenovo’s newly announced AI Services Center Of Excellence (COE) combines business advisors, data scientists, and AI-optimized infrastructure as-a-service to maximize customers’ strategic AI outcomes. Lenovo’s new AI professional services offerings include:

  • AI Discover – Helps customers uncover the “art of the possible” for AI with workshops and assessments, looking across the ecosystem and mapping out the blueprints for AI success.
  • Fast Start Generative AI services with NVIDIA – Lenovo offers full-stack solutions to support the complete product lifecycle, plus services to implement and scale Generative AI solutions.
  • TruScale GenAI As-A-Service – Provides AI capabilities through an as-a-service model that increases flexibility, scalability and predictability.
  • Enhanced Professional Services for AI – Helps customers accelerate their AI transformation by providing business advisors, data scientists, and AI-optimized infrastructure-as-a-service.

For more information on how Lenovo is working with NVIDIA toward a smarter, faster future, visit https://www.lenovo.com/us/en/servers-storage/alliance/nvidia/.

Speak Your Mind