Supermicro Volume Shipments of Performance Servers Optimized for AI, HPC, Virtualization, and Edge Workloads

Supermicro, Inc. is commencing shipments of max-performance servers featuring Intel Xeon 6900 series processors with P-cores.

These systems feature a range of new and upgraded technologies with new architectures optimized for the most demanding high-performance workloads including large-scale AI, cluster-scale HPC, and environments where a maximum number of GPUs are needed, such as collaborative design and media distribution.

“The systems now shipping in volume promise to unlock new capabilities and levels of performance for our customers around the world, featuring low latency, maximum I/O expansion providing high throughput with 256 performance cores per system, 12 memory channels per CPU with MRDIMM support, and high performance EDSFF storage options,” said Charles Liang, president and CEO, Supermicro. “We are able to ship our complete range of servers with these new application-optimized technologies thanks to our Server Building Block Solutions design methodology. With our global capacity to ship solutions at any scale, and in-house developed liquid cooling solutions providing unrivaled cooling efficiency, Supermicro is leading the industry into a new era of maximum performance computing.”

There are currently several X14 systems available to customers for remote testing and validation via Supermicro’s JumpStart program.

X14 systems are available in variety of form factors, each optimized for range of performance-intensive workloads:

• GPU-optimized supporting the latest-gen of SXM and PCIe GPUs, featuring enhanced thermal capacities and direct-to-chip liquid cooling on some models.
• High density multi-nodes including all-new FlexTwin and GrandTwin models as well as the proven, award-winning SuperBlade architecture.

These models leverage shared components to increase efficiency and can be fitted with direct-to-chip liquid cooling for maximum performance density.

• Supermicro Hyper rackmounts combine single or dual socket architectures with flexible I/O and storage configurations in traditional rackmount form factors to help enterprises and data centers scale up and out as their workloads evolve.

The company’s max-performance X14 systems support the Xeon 6900 series processors with P-cores, which feature up to 128 performance cores/CPU, support for high bandwidth MRDIMMs up to 8,800MT/s, and built-in accelerators including the AI-specific Intel AMX.

The X14 systems represent a perfect building blocks for data centers at any scale, with the company able to provide complete rack-level integration services including design, building, testing, validation, and delivery. An industry-leading global manufacturing capacity of up to 5,000 racks/month (2,000 liquid cooled) and extensive testing and burn-in facilities allow Supermicro to deliver solutions at any scale in a matter of weeks, not months. With the firm’s complete in-house liquid-cooled direct-to-chip cold plate solutions, liquid cooling can be included in rack-level integrations to further increase system efficiency, reduce instances of thermal throttling, and lower both the TCO and Total Cost to Environment (TCE) of data center deployments. These turn-key solutions include the rack, cabling, power, and cooling infrastructure to simplify solution deployment at scale.

To maximize the performance and density potential of the latest X14 systems, the company also offers complete in-house developed liquid cooling solutions including cold plates for CPUs, GPUs, memory, cooling distribution units, cooling distribution manifolds, hoses, connectors, and cooling towers. Liquid cooling is easily included in rack-level integrations to increase system efficiency, reduce instances of thermal throttling, and lowers both the TCO and Total Cost to the Environment (TCE) of data center deployments.

Supermicro max-performance X14 systems featuring Xeon 6900 series processors with P-cores include:

• GPU-optimized – The highest performance the firm’s X14 systems designed for large-scale AI training, large language models (LLMs), GenAI and HPC, and supporting 8 of the latest-gen SXM5 and SXM6 GPUs. These systems are available in air-cooled or liquid-cooled configurations.
• PCIe GPU – Designed for maximum GPU flexibility, supporting up to 10 double-width PCIe 5.0 accelerator cards in a thermally-optimized 5U chassis or edge-optimized 3U chassis. These servers are for AI inferencing, media, collaborative design, simulation, cloud gaming, and virtualization workloads.
• Intel Gaudi 3 AI Accelerators – The company is now shipping an industry’s first AI server based on the Intel Gaudi 3 accelerator hosted by Xeon 6 processors. Designed to increase the efficiency and lower the cost of large-scale AI model training and AI inferencing, the system features 8xIntel Gaudi 3 accelerators on an OAM universal baseboard, 6xintegrated OSFP ports for cost-effective scale-out networking, and an open platform designed to use a community-based, open-source software stack, requiring no software licensing costs.
• SuperBlade – The firm‘s X14 6U high-performance, density-optimized, and energy-efficient SuperBlade maximizes rack density, with up to 100 servers and 200 GPUs per rack. Optimized for AI, HPC, and other compute-intensive workloads, each node features air cooling or direct-to-chip liquid cooling to maximize efficiency and achieve the lowest PUE with the best TCO, as well as connectivity up to 4 integrated Ethernet switches with 100G uplinks and front I/O supporting a range of flexible networking options up to 400G InfiniBand or 400G Ethernet/node.
• FlexTwin – The X14 FlexTwin architecture is purpose-built for HPC, cost-efficient, and designed to provide maximum compute power and density in a multi-node configuration with up to 24,576 performance cores in a 48U rack. Optimized for HPC and other compute-intensive workloads, each node features direct-to-chip liquid cooling only to maximize efficiency and reduce instances of CPU thermal throttling, as well as HPC low latency front and rear I/O supporting a range of flexible networking options up to 400G/node.
• Hyper – X14 Hyper is the company‘s flagship rackmount platform designed to deliver the highest performance for demanding AI, HPC, and enterprise applications, with single or dual socket configurations supporting double-width PCIe GPUs for maximum workload acceleration. Both air cooling and direct-to-chip liquid cooling models are available to facilitate the support of top-bin CPUs without thermal limitations and reduce data center cooling costs while also increasing efficiency.