Data center News
Lenovo, Nvidia To Deliver Joint Hybrid AI Solutions To Spur GenAI Adoption
A Lenovo executive tells CRN that the expanded alliance with Nvidia is a ‘massive opportunity’ for channel partners because it will help spur adoption of generative AI by businesses who don’t want all their data in the cloud by providing hybrid solutions that let them customize and run AI workloads in on-premises servers and even PCs.
Lenovo and Nvidia plan to jointly deliver hybrid AI solutions that will allow channel partners to “deliver AI to more business than ever before” as part of an expanded alliance.
The expanded agreement was made Tuesday by Lenovo CEO Yuanqing Yang and Nvidia CEO Jensen Huang Tuesday at the Lenovo Tech World 2023 event in Austin, Texas.
A key motivator for the new collaboration is to spur adoption of generative AI by businesses who don’t want all their data in the cloud by providing solutions that let them customize and run AI workloads in on-premises servers and even PCs, Lenovo and Nvidia executives told CRN.
“Creating private models on-site with your own data, that’s an even bigger opportunity, in our opinion, that’s really going to extend the value of what generative AI is all about,” said Scott Tease, vice president and general manager of high-performance computing and AI at Lenovo.
Tease called the Lenovo-Nvidia hybrid AI partnership a “massive opportunity” for channel partners because they are familiar with each customer’s business needs.
“Once those foundational models are available, they’re licensable, they’re ready to be personalized and privatized with their own data. The opportunity for that is massive. And the channel is going to play a huge role in that because they’re our customers’ experts,” he said.
“[Partners] can help us link up the technical capability to the business outcomes that our customers are looking for,” Tease added.
The Hybrid Solutions In The Expanded Lenovo-Nvidia Alliance
The hybrid AI solutions provided by Lenovo and Nvidia will allow businesses to customize general-purpose foundation models using Nvidia’s AI Foundations cloud service and then run them in Lenovo servers and workstations powered by Nvidia software and hardware.
On the hardware side, the initial solutions will include the Lenovo ThinkSystem SR675 V3 server, which comes with Nvidia L40S GPUs, Nvidia BlueField-3 data processing units and Nvidia Spectrum-X Ethernet networking. It will also include the Lenovo ThinkStation PX workstation PC, which comes with up to four Nvidia RTX 6000 Ada Lovelace GPUs in a desktop form factor.
The systems will expand next spring to include new Lenovo servers that use Nvidia’s MGX modular reference designs, which are meant to give OEMs like Lenovo quicker ways to deliver Nvidia-powered systems with a variety of configurations tuned to each customer’s needs.
“As chips get faster, networking gets faster, and systems get faster, we can’t have the traditional turnaround time of everything being a motherboard-down design,” said Bob Pette, vice president of enterprise platforms and professional visualization at Nvidia.
“So MGX was meant to be a way to make it easy to either upgrade I/O or acceleration with GPUs or upgrade to the latest Intel or AMD CPUs or Nvidia Grace CPU,” he added.
As for software and services, these systems will come equipped with Nvidia AI Enterprise, a software suite that includes the AI frameworks, pre-trained models and tools businesses need to develop and run AI workloads. The suite also includes the Nvidia NeMo framework, for building, customizing and deploying large language models (LLMs) for a variety of generative AI use cases.
Lenovo’s AI systems will also support the recently announced VMware Private AI Foundation with Nvidia platform, which will help businesses run generative AI applications on VMware Cloud Foundation when it becomes available next year.
Tease said these solutions will allow Lenovo, Nvidia and their channel partners “deliver AI to more businesses and ever before.”
“Our joint belief is that for AI to fulfill the potential that it has, it needs to be everywhere,” he said.
Lenovo Partner Sees ‘Significant’ Generative AI Demand
Insight Enterprises, a top Lenovo and Nvidia partner, has seen “very significant” demand for generative AI solutions from customers, according to Megan Amdahl, senior vice president of client experience and North America COO for the Chandler, Ariz.-based solution provider powerhouse.
“The amount of demand this is going to put on compute and storage is amazing, and so we’re just trying to get ahead of it with our messaging,” she said.
That messaging includes the anticipation that increased AI capabilities in PCs will drive a new wave of refresh opportunities in concert with the 2025 end-of-life for Windows 10 support.
To deliver effective generative AI solutions, Insight is working to ensure it can develop compelling internal uses cases, which includes Insight GPT, an internal platform that gives employees an “exhaustive bank of information at their fingertips,” according to Amdahl.
Amdahl said her team is incentivizing employees to develop generative AI use cases by giving away gifts such as Apple AirPods. She added that one prize in the works is a one-year lease for a Tesla electric car for whoever delivers the best client experience.
“That’s how into it Insight is. We really want to put distance between us and our competition on that front,” she said.