Lenovo has announced a new series of enterprise servers and solutions specifically designed for AI inferencing workloads. These consist of three different ThinkSystem and ThinkEdge servers.
“Enterprises today need AI that can turn massive amounts of data into insight the moment it’s created,” said Ashley Gorakhpurwalla, President of Lenovo Infrastructure Solutions Group. “With Lenovo’s new inferencing-optimized infrastructure, we are giving customers that real-time advantage.”
The first server in the new series, the ThinkSystem SR675i, is designed for the most demanding workloads in manufacturing, healthcare, and financial services. The system can run full LLMs with massive scalability.
The ThinkSystem SR650i, in turn, offers AI inferencing with high-density GPU compute and is easy to deploy in existing data centers. For edge environments, there is the ThinkEdge SE455i, a compact server for retail, telecom, and industrial environments. It can operate in temperatures between -5°C and 55°C.
The servers run on AMD EPYC processors and Nvidia accelerated computing. Lenovo Neptune’s liquid cooling can dissipate up to 100 percent of the heat, which should improve energy efficiency for intensive AI workloads. The TruScale pay-as-you-go pricing model should lower the financial barrier.
Hybrid AI Factory as a blueprint
The hardware forms the basis of what Lenovo calls the Hybrid AI Factory, a modular framework for building and operating AI solutions at scale. The platform combines the new inferencing servers with storage, networking, software, and orchestration.
Lenovo presents three platforms: Hybrid AI Inferencing with ThinkAgile HX and Nutanix AI for virtualized environments, a platform with Red Hat AI for robust enterprise deployments, and a solution with Canonical Ubuntu Pro for rapid implementation and experimentation.
The approach is designed to reduce time to market and lower deployment risks. Organizations can choose the setup that suits their maturity and ambitions without compromising on performance or security.