3rd Gen Intel® Xeon® Scalable Processors

Discover the Benefits

  • Strong, capable platforms for the data-fueled enterprise.

  • Next-generation platform for cloud-optimized, 5G-ready networks, and next-generation virtual networks.

  • Breakthrough HPC and high-performance data analytics innovation.



Performance Made Flexible

We know that you must build for your needs of today while planning for unknown future demands. Flexible infrastructure is key, one that is optimized for multi-cloud and AI, serving many workloads, anywhere, at any time.

That’s why we’ve created a platform that delivers the flexibility you need, bringing AI everywhere, from edge to cloud, helping our customers achieve more. More performance. More efficiency. And more wonderful.

3rd Gen Intel® Xeon® Scalable processors offer a balanced architecture with built-in acceleration and advanced security capabilities, designed through decades of innovation for the most in-demand workload requirements.

Through partnership with the world’s software leaders and solution providers, 3rd Gen Intel Xeon Scalable processors are optimized for specific customer workload types and performance levels, all with the consistent, open Intel architecture you know and trust.

3rd Gen Intel Xeon Scalable Processors Are...

Optimized for cloud, enterprise, HPC, network, security, and IoT workloads with 8 to 40 powerful cores and a wide range of frequency, feature, and power levels.

Infused with Intel Crypto Acceleration, enhancing data protection and privacy by increasing the performance of encryption-intensive workloads including SSL web serving, 5G infrastructure, and VPN/firewalls, while reducing the performance impact of pervasive encryption.

The only data center CPU with built-in AI acceleration, end-to-end data science tools, and an ecosystem of smart solutions.

Engineered for the demands of cloud workloads and to fuel a wide range of XaaS environments.

Fueled by Intel SGX, which protects data and application code while in use from the edge to the data center and multi-tenant public cloud.

Outstanding Gen-on-Gen Performance

The new 3rd Gen Xeon Scalable processors are based on a balanced, efficient architecture that increases core performance, memory, and I/O bandwidth to accelerate diverse workloads from the data center to the intelligent edge.

  • Built-in workload acceleration features include Intel Deep Learning Boost, Intel Advanced Vector Extensions 512, and Intel Speed Select technology.
  • Available with up to 40 powerful cores.

Available Multi-Socket Scale to Propel Insights

Select 3rd Gen Intel Xeon Scalable processors support up to 28 cores per processor in four-and-eight-socket configurations, driving enhanced performance, throughput, and CPU frequencies compared to previous-gen processors.

  • Four-to-eight–socket 3rd Gen Intel Xeon Scalable processors deliver multi-socket core count density, with up to 224 cores per platform in an 8-socket configuration.
  • Up to six Intel UltraPath Interconnect (Intel UPI) channels increase platform scalability and improve inter-CPU bandwidth for I/O-intensive workloads versus the previous generation, offering an agile balance between improved throughput and energy efficiency.

1.46x average gen-on-gen performance improvement.1

Up to 1.60x higher memory bandwidth vs. prior gen.2

Up to 2.66x higher memory capacity vs. prior gen.3

Up to 1.33x more PCI Express lanes per processor vs. prior gen.4
Now with PCIe Gen 4

Technology Overview

Built-in workload and service acceleration.

Intel Deep Learning Boost (Intel DL Boost) acceleration is built-in specifically for the flexibility to run complex AI workloads on the same hardware as your existing workloads.

  • With int8 – available on all 3rd Gen Intel Xeon Scalable processors, Vector Neural Network Instructions (VNNI) enhance inference workloads by maximizing use of compute resources, improving cache utilization, and reducing potential bandwidth bottlenecks.
  • With bfloat16 – available on select 3rd Gen Intel Xeon Scalable processors, the industry’s first x86 support of Brain Floating Point 16-bit (blfoat16)5 brings enhanced artificial intelligence inference and training performance with Intel Deep Learning Boost.

Intel Advanced Vector Extensions 512 (Intel AVX-512)

Boosts performance and throughput for the most demanding computational tasks in applications such as modeling and simulation, data analytics and machine learning, data compression, visualization, and digital content creation. Take advantage of 3rd Gen Intel Xeon Scalable processors’ increased memory bandwidth, improved frequency management, and 2x the FMA (intrinsics for floating point fused multiply-add [FMA] operations)—now across Platinum Gold, and Silver SKUs. Compared to Intel AVX2, Intel AVX-512 is designed to enable greater performance than ever before.

Learn more at intel.com/avx512

Intel Speed Select Technology (Intel SST)

Is a powerful collection of capabilities giving you fine-grain control over CPU performance that can help to optimize total cost of ownership. With Intel SST, one server can do more.
Lots more.

  • 3rd Gen Intel Xeon Scalable processors offer enhanced Intel SST capabilities, featuring Intel SST – Base Frequency (Intel SST-BF), Intel SST – Core Power (Intel SST-CP), and Intel SST – Turbo Frequency (Intel SST-TF) capabilities across most Intel Xeon Platinum and Gold processors.
  • 3rd Gen Intel Xeon Scalable processors, Y SKUs, support the new Intel SST – Performance Profile 2.0 (Intel SST-PP), with additional core count, frequency, profile, and power configuration opportunities.

Learn more at intel.com/speedselect


Intel Software Guard Extensions (Intel SGX) provides fine grain data and privacy protection via application isolation in memory, independent of operating system or hardware configuration.6 7

Learn more at intel.com/sgx

Intel Crypto Acceleration brings built-in Vector AES-NI, Vector CLMUL, Intel Secure Hash Algorithm Extensions, VPMADD52 instructions, and RSA/DH encryption protocols.5

Learn more at newsroom.intel.com/crypto

Intel QuickAssist Technology provides platformbased hardware acceleration for cryptography and data compression.

Learn more at intel.com/quickassist

Intel Total Memory Encryption delivers full physical memory encryption support to enhance data and VM protection.8 9

Intel Platform Firmware Resilience (Intel PFR) uses an Intel FPGA to protect, detect, and correct, providing NIST SP800-193 compliant firmware resiliency. Now, platform firmware can be validated before execution, while runtime monitoring and filtering help protect against manipulation. In case of a compromise, Intel PFR also provides automated recovery in minutes.


Intel Resource Director Technology (Intel RDT) delivers visibility and control over shared platform resources to optimize performance and help increase resource utilization. The latest 3rd Gen Intel Xeon Scalable processors bring improvements including the second gen of Memory.

Bandwidth Allocation (MBA 2.0), now featuring an advanced hardware controller with flexible and efficient bandwidth controls, and higher-resolution counters (32b) for Memory Bandwidth Monitoring.

Learn more at intel.com/intelrdt

Overview of 3rd Gen Intel Xeon Scalable Processors

Intel Xeon Platinum 8300 processors are the foundation for secure, agile, hybrid-cloud data centers. With enhanced hardware-based security and exceptional multi-socket processing performance, these processors are built for mission-critical, real-time analytics, machine learning, artificial intelligence, high performance computing (HPC), and multi-cloud workloads.10

With trusted, hardware-enhanced data service delivery, these CPUs deliver improvements in I/O, memory, storage, and network technologies to harness actionable insights from our increasingly data-fueled world. Designed for advanced analytics, artificial intelligence, and high-density infrastructure, Intel Xeon Platinum 8300 processors deliver exceptional performance, platform capabilities, and workload acceleration.

Learn more at intel.com/xeonscalable

With support for higher memory speeds, enhanced memory capacity, and up to four-socket scalability, Intel Xeon Gold 6300 and 5300 processors deliver improved performance, enhanced memory capabilities, hardware-enhanced security, and workload acceleration.11 These processors are optimized for demanding mainstream data center, multi-cloud compute, and network and storage workloads. With up to four-socket scalability, they are suitable for an expanded range of workloads.

Intel Xeon Silver 4300 processors deliver essential performance, improved memory speed, and power efficiency. They offer hardware-enhanced performance required for entry data center compute, network, and storage.

Up to 40 cores per Intel Xeon Platinum processor.

Up to 8 memory channels at up to 3200 MT/s.

Features built-in AI acceleration (Intel DL Boost) with int8 (VNNI) and bfloat165 quantization for accelerated AI inference and training performance.

Features Intel SGX, the most deployed trusted execution environment, with up to 512GB enclave per processor.

Up to 32 cores per Intel Xeon Gold processor.

Up to 8 memory channels at up to 3200 MT/s.

Featuring Intel SGX, the most deployed trusted execution environment, with standard SKUs supporting up to 64GB enclave per processor.

Up to 20 cores per Intel Xeon Silver processor.

Up to 8 memory channels at up to 2667 MT/s.

Featuring Intel SGX, the most deployed trusted execution environment, with up to 8GB enclave per processor.

Portfolio Products & Technologies

Intel® Optane™ persistent memory 200 series.

Intel® Optane™ persistent memory is a category of devices that deliver flexibility to the data center. Use it to augment memory capacity beyond what’s available on DRAM-only systems or to deploy a new persistent memory tier, delivering high performance and low latency for accelerated data processing.

Intel Optane persistent memory 200 series is Intel’s next-gen security-enabled, reliable, persistent memory module. It provides large capacity and native persistence to help extract more value from larger datasets and increase agility with fast access to more data closer to the CPU. Intel Optane persistent memory is supported by leading software providers, including SAP, Oracle, Microsoft, VMware, Nutanix, Citrix, Apache Spark, Aerospike, Redis, MemVerge, and more.

Four-socket 3rd Gen Intel Xeon Scalable platforms support in-memory database deployments fueled by Intel Optane PMem 200 series (App Direct mode only).

Learn more at intel.com/optanepersistentmemory

The latest Intel Optane persistent memory 200 series delivers.12:

Average of 32% higher memory bandwidth vs. prior gen & platform.

Up to 6TB of memory per socket for faster analysis of the largest datasets.

Up to 16 % lower power vs. prior gen.

Intel Optane SSD P5800X

Data creation continues to grow at exponential rates, and workloads are becoming ever more intense. For the hottest stored data, legacy storage options are increasingly a performance bottleneck, inhibiting the evolution of new architectures and application performance.

Caching and tiering the hottest data to higher-performance storage can break these bottlenecks. However, reliance on today’s NAND SSDs for write-intensive environments risks disk wear-out, which can increase maintenance costs and downtime. The Intel Optane SSD P5800X offers the ideal combination of no-compromise I/O performance and never-before-seen SSD endurance to deliver unprecedented storage value.

Learn more at intel.com/optane

Compared to the previous generation, the Intel Optane SSD P5800X enables unprecedented storage value by delivering up to13:

60% lower latency for faster actionable insights.

50% better Quality of Service (QoS) to enable monetization of improved SLAs.

4K greater random 4K mixed read/write IOPs to better saturate high-speed.

67% higher endurance to help extend the life of lower-endurance NAND SSDs.

Intel SSD D5-P5316 (NAND)

Architected with the industry’s first 144-layer quad-level cell (QLC) NAND and PCIe Gen 4, the Intel SSD D5-P5316 enables significant TCO savings while accelerating warm storage.

Built from the ground up to optimize and accelerate storage, the Intel SSD D5-P5316, with Intel’s most advanced QLC NAND technology, delivers industry-leading SSD storage density for the data center and the high bandwidth of PCIe Gen 4. Compared to HDDs, the Intel SSD D5-P5316 can accelerate access to stored data by up to 25x, while performance optimization brings up to 2x higher sequential read performance, up to 38% higher random read, and up to 5x higher endurance compared to previous-gen NAND-based Intel SSDs.14

The innovative form factor also enables up to 1 petabyte of storage in just 1U of rack space, allowing massive storage consolidation for lower TCO. In fact, it facilitates up to a 20x reduction in storage footprint in the data center.15 Well-suited for a wide range of read-intensive, low-latency workloads, the Intel SSD D5-P5316 is ideal for content delivery network (CDN), hyperconverged infrastructure (HCI), big data, artificial intelligence (AI), cloud elastic storage (CES), and HPC workloads.

Learn more at intel.com/3dnand


Intel Ethernet 800 series adapters now offer port data rates from 10 Gbps to 100 Gbps, supporting both PCIe Gen 3 and Gen 4 across a variety of port counts to meet the needs of almost every workload.

  • Our Application Device Queues (ADQ) feature prioritizes application traffic to help deliver the performance required for high-priority, network-intensive workloads.
  • Enhanced Dynamic Device Personalization (DDP) uses the fully programmable pipeline to enable frame classification for advanced and proprietary protocols on the adapter—to increase throughput, lower latency, and reduce host CPU overhead.
  • Supports RDMA over iWARP or RoCEv2 protocols and NVMe over TCP with ADQ for high-throughput, low-latency cloud, storage, and HPC clusters.

The newest member of the adapter family, the Intel Ethernet Network Adapter E810-2CQDA2 increases network data throughput up to 200 Gbps per adapter for high-performance vRAN, NFV forwarding plane, storage, HPC, cloud, and CDN.

Learn more at intel.com/ethernet

Intel QuickAssist Technology

Intel QuickAssist Technology (Intel QAT) offloads CPU to dedicated hardware acceleration for:

  • Cryptography
  • Data compression

Available in two form factors in all generations of Intel Xeon Scalable processors to date, Intel QAT is already widely deployed and proven.

  • External Intel QuickAssist Adapters as PCIe cards
  • External integrated in the PCH as chipset

Enhance security and compression performance in cloud, networking, big data, and storage applications— for data in motion and at rest. Now, you can accelerate compute-intensive operations with Intel QAT. That includes symmetric encryption and authentication, asymmetric encryption, digital signatures, RSA, DH, ECC, and lossless data compression.

Learn more at intel.com/quickassist

Customer Segment Benefits

5G performance made flexible.

5G use cases come in all shapes and sizes—so should your infrastructure. Workloads are unique and need to be delivered across multiple locations, meaning that as your network evolves, the technology must too.

Network-optimized 3rd Gen Intel Xeon Scalable processors, N SKUs, are designed to support diverse network environments. Optimized for many workloads and performance levels, they are available in a wide range of cores, frequencies, features, and power. For organizations ready to drive 5G networking to the next level, these CPUs increase 5G user plane function (UPF) performance by up to 1.42x vs. the prior gen.1

Through partnership with a proven, broad ecosystem, including 400+ Intel Network Builders members, Intel is delivering solution blueprints based on 3rd Gen Intel Xeon Scalable processors, N SKUs—all resulting in accelerated qualification, plus shortened time-to-deployment and value for vRAN, NFVI, virtual CDN, and more.

  • Network-optimized N SKUs deliver low latency and high throughput with energy efficiencies.
  • Enhanced Intel AVX-512 & Crypto processing.
  • Robust and proven ecosystem to ensure time to deployment.
  • Intel Crypto Acceleration enhances encrypted network workloads and services and virtually eliminates the performance impact of full data encryption for normal workloads, with dedicated Intel QAT hardware offload options available too.
  • With up to 36 latest-gen cores, these CPUs offer higher base frequency for greater throughput for virtualized network functions and lower power for dense or constrained physical deployments.
  • Accelerate networking infrastructure deployment time with our OEM partners, and leading cloud providers.
  • Multiple SKUs available with increased supply availability of up to 7 years.

Learn more at intel.com/networking

Your 5G advantage1

1.62x average performance improvement across network and communication workloads vs. prior gen.

2x massive MIMO throughput in a similar power envelope for a best-in-class 3x100MHz 64T64R configuration.

Upto 1.76x Up to enhanced DPDK L3 forwarding vs. prior gen.

1.63x Up to increased CDN throughput, enabling you to serve the same number of users at a higher resolution or more subscribers at the same resolution vs. prior gen.

AI Performance Made Flexible

Whether you’re making product recommendations, optimizing your supply chain, or sequencing genomes, AI is integral to your digital future. But delivering AI into production from edge to cloud has been challenging, with infrastructure considerations, software and application compatibility, and fragmented tools top of mind. Until now.

As the only data center CPU with built-in AI acceleration, hardware-enhanced security, and software optimizations, 3rd Gen Intel Xeon Scalable processors can help deliver higher performance per TCO across the most diverse set of smart AI workloads.

Through Intel’s optimizations to streamline popular end-to-end data science tools, powered by the open oneAPI standard, 3rd Gen Intel Xeon Scalable processors make it easier and faster for data practitioners to build and widely deploy smarter models—and simpler to move smoothly from PoC to production.

Choose from a broad selection of pre-integrated and verified enterprise solutions for data analytics and AI. Combined with the largest enterprise databases and a broad range of applications ready to unlock your insights, accelerate your time to AI deployment with 3rd Gen Intel Xeon Scalable processors.

Key Uses and Workloads

Classical machine learning
Natural language processing
Recommender system
High-res computer vision
Federated learning

  • Intel DL Boost delivers higher performance with built-in AI.
  • Intel SGX helps improve security with multi-party computing through federated learning.
  • Use built-in Intel Crypto Acceleration to virtually eliminate the performance impact of full data encryption and dramatically increase the performance of encryption-intensive workloads and/or Intel QAT in cases where dedicated hardware offload is beneficial.
  • Software optimizations on industry standard framework help boost performance.
  • M SKU is optimized for AVX-heavy vector instruction workloads, as found in media processing, AI, and HPC. In addition to enhanced user experiences, it also offers improved performance per watt.

Learn more at ai.intel.com

Your AI advantage1

1.74x Up to more AI inference performance with enhanced Intel Deep Learning Boost vs. prior gen.

Security Made Flexible

The connected world can expose valuable data assets to theft or tampering. Privacy concerns and regulations can make some collaboration opportunities impossible with today’s solutions, which protect data integrity in storage or on a network, but can leave data exposed while in use.

The latest 3rd Gen Intel Xeon Scalable processors— delivering a revolutionary step forward for privacy and security—include Intel SGX, which protects data and application code while in use, enabling new ways to collaborate using shared data without compromising privacy. With the smallest attack surface within the system, Intel SGX is already proven in countless research studies and production deployments by many leading security organizations.

  • Intel SGX creates secured enclaves in memory where data and application code can be verified, protected, and kept private while being processed.
  • Strengthen today’s security workloads, such as key management systems, and open new possibilities of trusted multi-party compute and collaboration while keeping each party’s data private.
  • Keep data and code isolated and better protected from intruders, malicious software, and even your cloud provider.

Key Uses and Workloads

Trusted foundation for all workloads
Cloud confidential computing
Multi-party compute
Enterprise blockchain
Key management

  • Built-in set of security capabilities address current and future privacy and security concerns customers face.
  • Intel SGX delivers the most proven and deployed Trusted Executive Environment (TEE) for the data center, now with huge memory enclaves.
  • New Intel Crypto Acceleration reduces CPU overhead for better encryption performance and SLA on encryption security.
  • Intel QAT has already been widely tested and deployed in the market—with improvements implemented over time.
  • New Intel Total Memory Encryption (Intel TME) encrypts full system memory for added protection against physical attacks.
  • Support for up to 512 GB Intel SGX enclave capacity, per CPU, available on multiple SKUs. See SKU table on Page 14 for more details.

Learn more at intel.com/sgx

Your security advantage1

1.48x Up to faster encryption performance with Intel Crypto Acceleration vs. prior gen.

Cloud and Enterprise Made Flexible

You benefit when your cloud infrastructure is flexible enough to serve many workloads. Deploying on an optimized platform tuned to your multi-cloud needs and delivered with broad ecosystem support can help you scale your solution.

Intel enables a simpler path and seamless migration to trusted cloud deployments and speeds the deployment of proven, repeatable cloud use cases in most industries, built from deep, varied industry expertise and collaboration.

The latest 3rd Gen Intel Xeon Scalable processors are built upon years of cloud innovation, tuned to your unique needs with built-in AI and Intel Crypto Acceleration and advanced security capabilities.

Built on open standards and APIs, with fully optimized software and security from hardware up the stack, 3rd Gen Intel Xeon Scalable processors scale up and out with ease.

As a result, you have flexibility to evolve your infrastructure for changing needs, streamlining cost and enhancing data governance.

All of the world’s largest cloud service providers plan to offer services powered by 3rd Gen Intel Xeon Scalable processors, ranging from IaaS, e-commerce, and content distribution, to social media, confidential computing, and more.

Key Uses and Workloads

Relational database
Web services
Virtual desktop (VDI, WVD)
Security (confidential compute)

  • Built-in acceleration technologies and advanced security to enable a safer remote workforce.
  • Accelerate deployment with ISV partnerships.
  • Interoperability is tailored to power mission-critical workloads with consistency across the ecosystem.
  • Intel delivers disruptive performance efficiency across workloads, agile, trusted infrastructure, and differentiated services to grow new markets.
  • Intel SGX helps deliver privacy and security in cloud solutions on the market today, like Microsoft Azure Confidential Compute, Alibaba Cloud, Baidu MesaTEE, and IBM Cloud Data Shield.

P SKU: IaaS optimized for orchestration efficiency in high-frequency VM environments.
V SKU: SaaS optimized SKU for orchestration efficiency in high-density, lower-power VM environments.

Y SKU: Intel Speed Select Technology – Performance Profile.

Learn more at intel.com/cloud

Your cloud & enterprise advantage1
Up to 1.58x higher performance on cloud-based microservices vs. prior gen.
Up to 1.72x higher virtualization performance vs. prior gen.
Up to 1.64x more database transactions per minute vs. prior gen.

High Performance Computing (HPC) Made Flexible

3rd Gen Intel Xeon Scalable processors are designed to power new business breakthroughs and scientific discoveries through outstanding performance and flexibility in an efficient, future-forward server infrastructure that supports the convergence of HPC with AI in the cloud or on-premises.

Enhanced core architecture, memory bandwidth, and security capabilities deliver faster time to results for HPC customers, while amazing performance, both gen-on-gen and versus competition, powers a wide range of diverse and challenging HPC applications.

New Q SKUs, optimized for liquid-cooled systems, provide HPC customers with max core count and frequencies delivering enhanced top-bin performance. Memory-bound workloads are fueled with faster I/O, featuring increased memory bandwidth, more capacity with 8 channels, and 64 lanes of PCIe Gen 4 per CPU.

Built-in advantages such as Intel DL Boost and enhanced Intel AVX-512 deliver AI and HPC convergence and workload performance, while new Intel Speed Select Technology capabilities empower HPC customers to tailor performance for their specific workload needs. See also the M SKU.

Key Uses and Workloads

Life sciences
Security (confidential compute)

  • Core count and frequency flexibility.
  • 8 DDR4 memory channels per CPU at up to 3200 MT/s.
  • Up to 64 lanes PCI Express Gen 4 per socket to enable higher I/O bandwidth per core.
  • Added improvements for performance per core and performance per watt.
  • Accelerate deployment time for HPC workloads with our OEM partners, and leading cloud providers.

Q SKU: Liquid cooled for high performance

Learn more at intel.com/hpc

Your HPC advantage1
1.53x Up to higher HPC performance vs. prior gen.

Internet of Things (IoT) Made Flexible

Compute-intensive workloads such as AI and video analytics are headed to the intelligent edge and IoT, where processing can take place closer to the data source or point of action. The need for performance, security, and manageability continues to grow as the diversity, sophistication, and intelligence of edge and IoT deployments accelerates.

The newest 3rd Gen Intel Xeon Scalable processors deliver the performance, security, and operational controls required for powerful AI, complex image or video analytics, and consolidated workloads at the edge, on-prem, or wherever work needs to be done.

This processor family offers enhanced gen-over-gen performance and built-in AI acceleration for faster analytics, processing of more images or video streams, and more powerful AI in edge and IoT deployments. Built-in hardwarebased security protects critical code and private data from tampering or interception by malware or hackers, a particular challenge in dispersed edge, industrial, or IoT deployments.

Key Uses and Workloads

Industrial manufacturing
Public sector
Retail, banking, hospitality, and education

  • Greater throughput and built-in AI acceleration for faster analytics, more video streams, and more powerful AI in edge and IoT.
  • Increased performance and more cores along with increased memory bandwidth3 allows for faster object recognition analysis on multiple video streams, simultaneously, for vision usages.
  • Built-in Intel Deep Learning Boost can accelerate deep learning workloads such as image classification, object detection and more. In healthcare deployment this can augment clinical workflows and help provide diagnosis.
  • Enhanced per-core performance enables convergence of workloads and more virtualized desktops onto single or fewer backend servers for retail, banking, and education deployments.
  • Long-life availability to meet the specific support needs of the intelligent edge, industrial, and embedded segment.

T SKU = Up to 10-year usability on select SKUs, support for higher case temperature (TCase)

Learn more at intel.com/iot

Your IoT advantage1
Up to 1.56x improvement in AI inference for image classification with enhanced Intel DL Boost vs. prior gen.

Hardened security and long-life availability to meet the specific needs of edge and embedded systems.