Skip to main content

Supermicro Expands Data Center Optimized Total IT Solutions with 4th Gen AMD EPYC™ Processors, Delivering World Record Performance for Today’s Most Critical Workloads

All New Single and Dual Socket Servers for Cloud, AI/ML, HPC, HCI, and Enterprise Applications – Contain Up to 192 Cores, up to 12TB DDR5-4800MHz 12-channel Memory, and up to 160 PCIe 5.0 Lanes to Drive the Most Advanced Applications

San Jose, Calif., November 10, 2022 – Supermicro (NASDAQ: SMCI), a Total IT Solution Provider for Cloud, AI/ML, Storage, and 5G/Edge, is announcing significant additions to its already broad product line with servers supporting the new 4th Gen AMD EPYC processors. These innovative Supermicro systems will transform how enterprises analyze massive amounts of data, perform complex simulations, and can lower the total cost of ownership (TCO). With the new servers based on the 4th Gen AMD EPYC processors, eight new world records have been established, which brings the total to over 50 world records previously set.

Supermicro servers are designed to support the fastest and highest-core AMD EPYC 9004 Series processors, each with up to 96 cores and 3TB of memory per CPU. Up to 160 PCIe 5.0 lanes are available in dual processor systems, with eight bonus lanes of PCIe 3.0 or 4.0 for less demanding peripheral connections. All 4th Gen AMD EPYC processors include a set of state-of-the-art security features that help keep data secure, whether in use, inflight, or in-store.

“Supermicro is again leading the industry with some of the first-to-market new systems that deliver Total IT Solutions based on the new AMD EPYC 9004 Series processors,” said Charles Liang, president, and CEO, of Supermicro. “Our building block architecture gives us a first-to-market advantage and enables Supermicro to deliver application-optimized servers to meet the growing demands of our customers. We are incorporating the latest technologies into various form factors and platforms, including 96-core CPUs, DDR5 memory, PCIe 5.0 storage, networking, accelerators, and CXL 1.1+ peripherals. Our highly configurable systems are designed for the highest TDP CPUs that can be liquid cooled, greatly increase performance per watt, tailored for customer’s specific requirements, and have a reduced TCO. Our high-performing GPU servers, ideal for AI applications, support NVIDIA H100 Tensor Core GPUs and are perfect for metaverse-type applications.”

The new Supermicro server product lineup includes the following:

  • GrandTwin™ –The Supermicro GrandTwin is the newest multi-node architecture solution with font and rear I/O, designed for maximum density and purpose-built for single-processor performance per node on the latest AMD EPYC 9004 Series processors, and contains 12 DIMM slots. The flexible modular design is optimized for a wide range of applications, with a front I/O option fully accessible from the cold aisle, simplifying installations and servicing space-constrained environments.
  • CloudDC – The Supermicro CloudDC servers are single-processor systems optimized for I/O flexibility and 12 DIMM slots for many cloud-focused applications. Offering convenient serviceability with toolless brackets, hot-swap drive trays, and redundant power supplies result in rapid deployment and efficient maintenance in data centers. They are designed for cost-effective service delivery in cloud computing, including web hosting, email services, public and private cloud computing, and content-delivery networks.
  • Hyper – The Supermicro Hyper solutions are enterprise-focused servers built with versatility and performance. Uncompromised performance design with dual-processors and 12 channel 24 DIMMs optimized for supporting the highest TDPs and offering a flexible range of computing, networking, storage, and I/O expansion capabilities.
  • 4U GPU Servers – The Supermicro 4U GPU servers are 4U dual-processor systems, supporting up to 10 FHFL double-width PCIe GPU cards, including the latest AMD Instinct™ MI200 Series and NVIDIA H100 GPUs. The 4U GPU-optimized systems provide maximum acceleration, flexibility, and balance for AI, deep learning, and HPC applications and contain 24 DIMMs.
  • 8U GPU Servers – The Supermicro 8U universal GPU servers are the next-generation machine learning platform with 8x NVIDIA HGX H100 accelerators and dual processors, with 24 DIMMs. Flexible configurations, dual zone cooling, and end-to-end non-blocking bandwidth to GPU provide the highest utilization and lowest TCO and training cycles.

“Building on the record-breaking performance of 3rd Gen AMD EPYC processors, the latest 4th Gen AMD EPYC processors help our customers achieve better business outcomes faster and address their most ambitious energy efficiency goals. Our new “Zen 4” architecture is optimized for modern workloads and delivers the core density, memory bandwidth, and sophisticated security features customers demand,” said Ram Peddibhotla, corporate vice president of EPYC product management, AMD.

“Astera Labs is collaborating with Supermicro to make CXL a reality in data-centric platforms and solve performance and TCO challenges in memory-intensive workloads,” stated Jitendra Mohan, CEO, Astera Labs. “We look forward to delivering industry-leading solutions with Supermicro that leverage our Leo Memory Connectivity Platform and 4th Gen AMD EPYC processors to unlock cloud server performance with CXL technology.”

Supermicro will also make available several Supermicro servers with 4th Gen AMD EPYC™ 9004 Series processors through its popular JumpStart program. Customers may apply for access at: www.supermicro.com/jumpstart/h13.