Phone-to-Cloud AI Inferencing from Lenovo and Intel®
Deliver powerful AI solutions—efficiently—using Lenovo ThinkSystem Servers with Intel® Xeon® 6 processors
Artificial intelligence (AI) is not one-size-fits-all. It’s become essential to virtually every business, large and small. It’s evolving quickly, with a steady flow of new solutions using advanced AI inferencing, small language models (versus large ones), and more. Plus, deploying large-scale AI solutions—all the way to our personal smartphones, in many cases—can cost a lot, both in compute capacity and the energy to keep it all running.
To more rapidly deliver on the promise of AI, at a cost more companies can afford, today’s leading solutions are adopting a purpose-built approach that optimizes performance and efficiency. And to power these systems, AI architects are turning to Lenovo ThinkSystem servers and ThinkEdge devices powered by Intel® Xeon® technology, including Intel® Xeon® 6 processors. Combined with the available Intel® Distribution of the OpenVINO™ toolkit, designed to accelerate deep learning model inference tasks, it makes a robust platform for AI deployment across data centers, edge locations, and hybrid environments.
Together, Lenovo and Intel® (and AI) are empowering companies to deploy new, efficient, fit-for-purpose industrial and customer solutions faster than ever before. Keep reading to learn more.
Lenovo, Intel® and AI Inference Servers
As AI is used in more aspects business and technology, it’s good to think of the “magic” as happening in two distinct phases. It starts with training, where foundational AI models are developed using curated datasets. Then comes AI inferencing, where the trained models are applied to fresh data to answer new questions or perform new tasks. As we’ve all heard, the first part, AI training, requires enormous computational resources, limiting it to the biggest enterprises and organizations. However, today, many companies are realizing that the inference part of AI, with its reduced processing demands, can be more broadly—and less expensively—adopted by businesses of almost any size.
Plus, because AI inference tasks analyzing new data are less resource-intensive than the initial training process, lots of inference processing can be performed at the “edge”—that is, where the fresh data is being collected through remote sensors, video cameras, user’s smartphones, etc. So in a manufacturing plant, inference edge servers might use trained AI models to identify defects and take corrective actions—without the latency or cost of constantly sending the data back and forth with a cloud-based server.
For example, Lenovo ThinkSystem V4 Neptune servers, powered by Intel® Xeon® 6 processors, are built to excel in edge AI applications. Their power-efficient performance and reduced cooling demands make them suitable to deploy just about anywhere, including existing data centers. The Intel® Xeon® 6 processors deliver more performance per watt than earlier designs, and Lenovo Truscale metering technology and energy management tools can help further optimize power usage and related concerns.
Here's a closer look at four Intel® Xeon®-powered Lenovo ThinkSystem servers*:
ThinkSystem SR630 V4 server:
ThinkSystem SR650a V4 server:
ThinkSystem SR650 V4 server:
ThinkSystem SC750 V4 Neptune server:
- A supercomputer designed for exascale to enterprise-scale applications
- Up to two Intel® Xeon® 6900-series processors with high-bandwidth MRDIMMs
- Features Lenovo Neptune direct water cooling for performance and reliability
- Support for advanced connectivity options including 400 GB Ethernet
*In-stock availability of these products will vary over time. It you’re shopping, use the navigation options on the left side of this webpage to view today’s full line-up of Intel-powered servers from Lenovo.
Real-World Results with Lenovo, Intel® and AI Inference Devices
Today, AI has evolved far beyond the “theoretical” stage. It’s now delivering true, real-world results for companies large and small. And Lenovo ThinkSystem servers with Intel® Xeon® processors are playing a big role in pushing AI inference processing to the edge, whether you’re monitoring a far-off manufacturing plant or offering services over user smartphones. Here are brief recaps of three solution briefs we’ve published (see links higher on this page):
- AI-assisted visual inspections in industry: An industrial software company enhanced data analytics and AI-assisted visual inspections in Industry 4.0 scenarios by using its software with Lenovo edge devices and the OpenVINO™ toolkit from Intel®.
- Maximizing oil & gas well efficiency: Intel®-powered Lenovo edge products help drive an AI company’s oil and gas well monitoring solution—in one scenario it tries to predict pump failures and other events to increase the lifespan of critical equipment.
- Real-time data insights from the edge: Lenovo teamed with an edge-focused AI company on a solution that optimized the company’s algorithms for the OpenVINO™ toolkit from Intel® to extract real-time, actionable data that clients can use to save money or grow revenue.
Watch video
