Super Micro Computer, Inc., a Total IT Solution Provider for AI, Cloud, Storage, and 5G/Edge, announces the launch of a new, versatile, high-density infrastructure platform optimized for AI inferencing at the network edge. As companies seek to embrace complex large language models (LLM) in their daily operations, there is a need for new hardware capable of inferencing high volumes of data in edge locations with minimal latency. Supermicro's innovative system combines versatility, performance, and thermal efficiency to deliver up to 10 double-width GPUs in a single system capable of running in traditional air-cooled environments.

"Owing to the system's optimized thermal design, Supermicro can deliver all this performance in a high-density 3U 20 PCIe system with 256 cores that can be deployed in edge data centers," said Charles Liang, president and CEO of Supermicro. "As the AI market is growing exponentially, customers need a powerful, versatile solution to inference data to run LLM-based applications on-premises, close to where the data is generated. Our new 3U Edge AI system enables them to run innovative solutions with minimal latency."The new SYS-322GB-NR includes two powerful Intel Xeon 6900 processors with P-cores, 8800 MT/s MRDIMM and up to 20 PCIe 5.0 expansion slots. This Supermicro system supports a variety of single or double-width GPUs, or to use some of the expansion slots for high-performance I/O or other add-on cards. Additionally, the server features up to 6 TB of RDIMM memory and up to 14 E1.S or 6 U.2 NVMe drives.

One example use case that this system delivers is in the manufacturing industry, where Supermicro's new system can be deployed on-site at an automated production environment to process data feeds from cameras and sensors without having to transfer the data to a remote location. This capability reduces networking requirements and improves response times. Another environment where the SYS-322GB-NR will excel is large-scale control rooms, where the AI accelerator cards can be partially replaced by multi-display cards to support up to 64 independent displays.

Supermicro at Mobile World Congress (MWC) Las Vegas

The SYS-322GB-NR will be on display at Supermicro's booth #518 during MWC Las Vegas, October 8-10. Additionally, Supermicro will display systems that incorporate NVIDIA, AMD and Intel Xeon 6 processors, including X14 family edge and telco systems such as:

SYS-222HE-FTN - the Hyper-E brings data center performance to the telco edge with a dual Intel Xeon 6 processor in a 2U, short-depth form factor with front I/O access

SYS-212B-FN2T - a 2U short-depth system for AI in telco and edge deployments, featuring a single Intel Xeon 6700 series processor with E-cores and GPU support

SYS-E403-14B-FRN2T - a box PC sized, wall-mountable edge device capable of bringing the Intel Xeon 6700 series processor with E-cores and GPU support to remote environments

AS -1115S-FDWTRT - a 1U NEBS compliant system providing Telco performance for ORAN, Core and managed services. This system utilizes the AMD EPYC 8004 Series processor and support of up to 1 single-width GPU accelerator for heavy workloads.

In addition to showcasing Supermicro hardware systems, in partnership with NVIDIA, we will jointly demonstrate inferencing and AI solutions for on-premise and Edge applications including Enterprise AI, Retail, Telco Edge and Financial Services. We will demonstrate key generative AI solutions that include NVIDIA NIM, NVIDIA NeMo, NVIDIA Metropolis, remote management, security, and networking. For Telecom, Supermicro and NVIDIA will be demonstrating a live AI RAN solution using NVIDIA and Supermicro solutions showcasing performance, management and AI use cases.

Also, on display at MWC Las Vegas is the new a joint solution from Supermicro and Intel which combines the ruggedized IP65 Outdoor Edge system with a built-in AI Network Accelerator and Intel Data Center GPU Flex 170. This solution enables fast and cost-effective deployment of multiple private 5G networks as well as Edge AI applications in a single device. The networks can be used and exploited by different users, offering a scalable solution for dense environments such as industrial and campus sites, venues and smart cities.