Supermicro Expands NVIDIA Ampere Architecture-Based GPU Line For Enterprise AI With Industry-First 5 PetaFLOPS 4U Tier 1 AI Platform

Supermicro Expands NVIDIA Ampere Architecture-Based GPU Line for Enterprise AI with Industry-First 5 PetaFLOPS 4U Tier 1 AI Platform

Super Micro Computer, Inc. (Nasdaq: SMCI ), a global leader in enterprise computing, storage, networking and green computing , expands its market leadership in the growing areas of Artificial Intelligence (AI), Machine Learning (ML), High Performance Computing (HPC), 3D graphics visualization and desktop virtualization of work.

Supermicro has the broadest portfolio of Tier 1 systems that incorporate industry-leading capabilities to achieve 5 petaFLOPS of AI performance in a 4U form factor with the latest  NVIDIA A100, NVIDIA  A40 ,  NVIDIA  RTX A6000, NVIDIA  A10 , and NVIDIA A16.

“Our collaboration with NVIDIA allows us to design a comprehensive portfolio of GPU systems and platforms that offer more choice to customers. Customers can exactly match their workload requirements to the system architecture to achieve superior size, performance, efficiency and cost, ”says Charles Liang, president and CEO of Supermicro. “Supermicro offers optimized multi-GPU systems that provide advanced solutions for AI, Deep Learning, HPC and video streaming applications.

By leveraging our modular solutions and our time-to-market advantage, we continue to develop and market solutions optimized for applications. Our flexible 2U 2-node GPU system with PCIe 4.0 is designed for cloud gaming and social networking, and is industry-leading supporting up to three double-wide or six single-wide GPUs per node. ”

The Supermicro systems scope includes the flagship Intel-based SYS-420GP-TNAR or AMD-based AS -4124GO-NART (+), with the 4U server powered by the NVIDIA HGX A100 8-GPU card with an option 40 GB (HBM2) or 80 GB (HBM2e) of memory on each GPU. These systems are already available worldwide.

“From drug discovery designed entirely through artificial intelligence to designing modern, high-performance factories, high-performance computing is essential to powering the most innovative work in the world today,” said Justin Boitano , vice president and general manager of advanced and enterprise computing systems at NVIDIA.

“Supermicro’s NVIDIA Certified Systems provide customers with a wide range of servers designed to deliver optimal performance for AI, graphics and simulation workloads running in the data center, cloud and edge. ”

The new systems can run multiple advanced application workloads, including the NVIDIA AI Enterprise software suite, which is optimized and certified by NVIDIA to run on VMware vSphere and NVIDIA Omniverse Enterprise, a comprehensive platform for collaboration and simulation in real time that brings transformative 3D workflows to businesses of all sizes. Supermicro is showcasing these cutting-edge performance solutions at this week’s NVIDIA GTC21 conference.

From Edge to Cloud, AI, ML, HPC and 3D visualization are becoming essential technologies needed by all businesses and in all industries. The development of systems and the offer of platforms integrating the latest hardware and software allow a whole series of companies to become quickly productive and to implement their latest solutions, in different formats.

Supermicro is a major participant in the NVIDIA Certified Systems Program, which validates the hardware and software configurations necessary for maximum productivity. These systems can process huge amounts of data through advanced networking, allowing complex models to be developed and new algorithms to be integrated.

Read an Exclusive Interview: Understanding Consumer Identity and Access Management (CIAM)

AI and other sophisticated algorithms. 3D visualization provides a better understanding of physical reality, while graphical virtualization allows IT administrators to consolidate systems infrastructure and centralize and secure their data.

Systems certified by NVIDIA allow quick and easy deployment in the data center. Supermicro is also at the forefront of the industry with NEBS Level 3 certified GPU servers for 5G and telecommunications customers. Systems certified by NVIDIA allow quick and easy deployment in the data center. Supermicro is also at the forefront of the industry with NEBS Level 3 certified GPU servers for 5G and telecommunications customers.

Systems certified by NVIDIA allow quick and easy deployment in the data center. Supermicro is also at the forefront of the industry with NEBS Level 3 certified GPU servers for 5G and telecommunications customers.

Consistently at the forefront of the market, Supermicro supports the latest hardware and software solutions from NVIDIA. The range of supported GPUs includes the NVIDIA A100 80GB, A100 40GB, A40, RTX A6000, A30, A10, and A16 GPUs, intended for AI inference and consumer computing, graphics and consumer video public, and virtual offices, all with AI technologies.

All NVIDIA products are available through a wide range of Supermicro systems that use 3rd generation Intel Xeon Scalable processors or 3rd generation AMD EPYC CPUs. Thanks to Supermicro’s design excellence, more GPUs fit into a smaller space (higher density) than other system vendors.

Supermicro systems equipped with NVIDIA HGX A100 modules (opt for 40 GB of HBM memory or 80 GB of HBMe memory) include:

  • AS -4124GO-NART / + – Dual 3rd Generation AMD EPYC processors with 8-GPU NVIDIA HGX A100
  • SYS-420GP-TNAR / + – Dual 3rd Generation Intel Xeon Scalable processors with 8-GPU NVIDIA HGX A100
  • AS -2124GQ-NART / + – Dual 3rd Generation AMD EPYC processors with NVIDIA HGX A100 4-GPUs
  • Supermicro systems with PCIe 4.0, supporting NVIDIA A100 / PCIe, A30, A16, A10, and RTX A6000 include:
  • AS -4124GS-TNR – Dual 3rd Generation AMD EPYC processors supporting up to 8 double-wide GPUs
  • SYS-420GP-TNR – Dual 3rd Generation Intel Xeon Scalable processors supporting up to 10 double-wide GPUs
  • AS -2114GT-DNR Multi-Node – Single 3rd Generation AMD EPYC processor supporting up to 6 single-wide GPUs or 3 double-wide GPUs per node
  • SuperBlade SBA-4119SG – Single 3rd Generation AMD EPYC processor supporting 1 double-wide GPU or 2 single-width GPUs per blade, with a maximum of 20 blades in an 8U enclosure
  • SYS-740GP-TNRT Workstation – Dual 3rd Generation Intel Xeon Scalable processors supporting up to 4 double-wide GPUs

Check Out The New Enterprisetalk Podcast. For more such updates follow us on Google News Enterprisetalk News.