Application Optimized Server based on 3rd Gen Intel® Xeon® Scalable Processor

Comprehensive server, storage and networking product lines optimized for IT, Data Center, Embedded, HPC and cloud computing that supports 3rd Gen Intel® Xeon® Scalable Processor.

  • Over 100 Building Block Optimized Designs
  • Maximum Processor, Memory and I/O Performance
  • Max Performance, High Volume Cloud, High Efficiency Multi-Node, Mainstream





    3rd Gen Intel Xeon Scalable Processor Brief:

    The 3rd Gen Intel Xeon Scalable processors are the latest CPUs designed for a range of workloads. From the edge to on-premises and cloud data centers, this new CPU offers outstanding performance on several benchmarks and includes new features that enable a new level of security and AI acceleration.

    With more cores at higher frequencies than previous generation of CPUs, the 3rd Gen Intel Xeon Scalable Processors enable new and innovative applications to be created and deployed from 5G/Telco applications to large-scale analytics.

    The increase in I/O performance due to PCI-E 4.0 allows faster communication to accelerator options. The amount of directly addressable memory is also increased, allowing more data to be kept in memory. Along with the Intel Optane 200 series, it blurs the distinction between memory and storage.



    Rackmount Server

    Leading Multi-node Architectures

    • Optimized for highest processor TDPs
    • Up to 22 hybrid NVMe
    • Up to 3 double width GPUs
    • PCI-E 4.0 support with 64 lanes per socket; Total 128 lanes
    • Dual 3rd Gen Intel® Xeon® Scalable processors up to 270W and 32 DIM slots for maximum memory capacity



    Twin Server

    High Performance & Flexibility Enterprise Applications Rackmount System

    • Highly configurable 2U 4-node and 2U 2-node systems
    • 3rd Gen Intel® Xeon® Scalable processors,2 per node, up to 270W TDP
    • All-hybrid hot-swappable drive bays - NVMe,SAS or SATA (2.5” or 3.5” drives) - Up to 12 NVMe drives per node
    • 16 DIMMs + 4 Intel Optane Persistent Memory 200 series per node
    • PCI-E 4.0 AIOM (OCP 3.0 compliant) networking - 1 per node



    GPU Server with PCI-E

    High Performance and Flexibility for AI/ML and HPC Applications

    • High density systems for double-width, full length PCI-E GPU
      • 1U: support up to 4 PCI-E GPU
      • 2U: supporting up to 6 PCI-E GPU
      • 4U: supporting up to 10 PCI-E GPU
    • Optimized for graphics and rendering applications
    • Dual socket Intel® Xeon® Scalable processors up to 270W
    • NVIDIA GPUs supported



    GPU Server with HGX

    High Performance and Flexibility for AI/ML and HPC Applications

    • Dense and scalable multi-GPU powerhouse supports the latest HGX A100 8 SXM4 GPU
    • Next generation of NVIDIA NVLink™, with double the GPU-to-GPU direct bandwidth, almost 10X higher than PCI-E 4.0
    • New NVIDIA NVSwitch that is 2X faster than the previous generation
    • Networking up to 200G, GPUDirect RDMA and GPUDirect Storage
    • AIOM slot (OCP 3.0 compliant) support
    • NVIDIA certified system



    Storage Server

    Application-optimized High-Performance Storage Solution

    • New generation top-loading server optimized for field serviceability and field replacement
    • PCI-E 4.0 storage controller with hardware RAID and IT mode
    • Tool-less hot-swappable drive bays supporting 3.5 and 2.5” media
    • Flexible mix of hybrid HDD and SSD drive bays for best performance and TCO
    • Superior pullout drive drawer design
    • Hot swappable nodes, expanders, drives, power supplies and fans


    Better Faster Greener
    Better Performance Per Watt and Per Dollar 40%-60% Better Performance on Cloud Workloads Reduced Environmental Impact & Lower TCO
    • Over 100 Building Block Optimized Designs
    • Maximum Processor, Memory and I/O Performance
    • Max Performance, High Volume Cloud, High Efficiency Multi-Node
    • Thermal Capacity Supports Highest Clock Speeds
    • Support for Full Memory Configuration and Bandwidth
    • Both Free Air and Water Cooled
    • Titanium-Level (96%) Power Supplies
    • OpenBMC, OCP v3.0 SFF Cards
    • New AIOM Cards Provide I/O Flexibility with OCP Superset
    • On Average 62% Better Performance on Network and 5G Workloads
    • Web (Crypto) Acceleration
    • DDR4-3200MHz
    • 1.6x Memory Bandwidth
    • 2.66x Memory Capacity
    • 15-20% Lower Power Costs with Optimized Shared Resource Designs
    • Enhanced Security with Hardware Root of Trust, Total Memory Encryption, Software Guard Extension
    • PCI-E 4.0
    • 2x I/O Bandwidth
    • Multi-Generation Infrastructure for up to 65% CAPEX Savings
    • New Web Management Interface
    • Tool-less Designs
    • Advanced Networking with 200G InfiniBand Switch, and up to 4x 25GbE Switches
    • Modular Upgrades for Maximum Performance and Efficiency
    • Select Component Refresh Reduces e-Waste


    With the new 3rd Gen Intel Xeon Scalable processors, applications will benefit from several innovations.

    • More cores – for applications that scale with the number of available cores, performance will increase.
    • More extensive memory access – with more memory that can be accessed on the main memory bus, applications will perform better without waiting for data to be retrieved from storage devices.
    • Faster memory access – with higher memory bandwidth, applications will execute faster, requiring less time to wait for critical data.
    • Faster communication – with PCI-E 4.0, applications can communicate with PCIe devices at twice the speed as before, resulting in overall application performance increases.
    • Interconnect between sockets – for applications that require socket-to-socket communication; the faster UPI channels will reduce execution time.
    • AI instructions – accelerate AI inferencing applications with Intel Deep Learning Boost.