2nd Gen Intel® Xeon® Scalable Processors Brief

Second Generation Intel® Xeon® Scalable Processors

Télécharger le PDF

The Foundation for Data-Centric Innovation

Empowering Transformation in a Data-Centric Era

Second Generation Intel® Xeon® Processor Scalable Family

Intel® Xeon® Platinum 9200 Processors and Intel® Xeon® Platinum 8200 Processors

Intel® Xeon® Gold 6200 Processors and Intel® Xeon® Gold 5200 Processors

Intel® Xeon® Silver 4200 Processors

Intel® Xeon® Bronze 3200 Processors

Across an evolving digital world, disruptive and emerging technology trends in business, industry, science, and entertainment increasingly impact the world's economies. By 2020, the success of half of the world's Global 2000 companies will depend on their abilities to create digitally enhanced products, services, and experiences,1 and large organizations expect to see an 80 percent increase in their digital revenues,2 all driven by advancements in technology and usage models they enable.

This global transformation is rapidly scaling the demands for flexible compute, networking, and storage. Future workloads will necessitate infrastructures that can seamlessly scale to support immediate responsiveness and widely diverse performance requirements. The exponential growth of data generation and consumption, the rapid expansion of cloud-scale computing, emerging 5G networks, and the extension of high performance computing (HPC) and artificial intelligence (AI) into new usages require that today's data centers and networks urgently evolve—or be left behind in a highly competitive environment. These demands are driving the architecture of modernized, future-ready data centers and networks that can quickly flex and scale.

The Intel® Xeon® Scalable processor family provides the foundation for a powerful data center platform that creates an evolutionary leap in agility and scalability. Disruptive by design, this innovative processor sets a new level of platform convergence and capabilities across compute, storage, memory, network, and security. Enterprises and cloud and communications service providers can now drive forward their most ambitious digital initiatives with a feature-rich, highly versatile platform.

Enabling Greater Efficiencies and Lower TCO
Across infrastructures, from enterprise to technical computing applications, the Intel® Xeon® Scalable processor family is designed for data center modernization to drive operational efficiencies that lead to improved total cost of ownership (TCO) and higher productivity for users. Systems built on the Intel® Xeon® Scalable processor family are designed to deliver agile services with enhanced performance and groundbreaking capabilities, compared to the prior generation.

Performance to Propel Insights

Intel's industry-leading, workload-optimized platform with built-in AI acceleration, provides the seamless performance foundation for the data-centric era from the multicloud to intelligent edge, and back, the Intel® Xeon® Scalable processor family with 2nd Gen Intel® Xeon® Scalable processors enables a new level of consistent, pervasive, and breakthrough performance.

New Intel® Xeon® Platinum 9200 processors

New 2nd Gen Intel® Xeon® Scalable processors

2nd Gen Intel® Xeon® Scalable Processor Provides Consistent Performance for Diverse Workloads

  What Matters? Intel® Xeon® Scalable Processor Family Advantages
Enterprise and Cloud

Minimize complexity with compatible virtualization infrastructure

Meet stringent customer SLAs

Deploy quickly. Intel VMs coexist w/ other Intel® technology-based servers.

Fast response times.
HPC
 Maximize vector floating point performance and efficiency High-performance with fewer servers.
Storage Ensure deterministic storage response Deterministic performance. Cores, cache, memory, I/O in a single die.
Communications Deliver diverse services efficiently Incredible efficiency and hardware acceleration from a platform with features required for application, control, packet, and signal processing.
Voir plus Voir moins View all

Support for Breakthrough Memory Innovation
A new foundation for performance begins with support for Intel's breakthrough Intel® Optane™ DC persistent memory, a new class of memory, and storage innovation architected for data-centric class environments. With individual module capacities up to 512 GB, Intel® Optane™ DC persistent memory can deliver up-to 36 TB of system-level memory capacity when combine with traditional DRAM. Intel® Optane™ DC persistent memory complements DRAM memory by affordably enabling unprecedented system memory capacity to accelerate workload processing and service delivery.

Learn more about Intel® Optane™ DC persistent memory

New 2nd Gen Intel® Xeon® Scalable Processors with Intel® Optane™ DC Persistent Memory

Foundational Enhancements

  • Higher Per-Core Performance: Up to 56 cores (9200 series) and up to 28 cores (8200 series), delivering high-performance and scalability for compute-intensive workloads across compute, storage, and network usages
  • Greater Memory Bandwidth/Capacity: Support for Intel® Optane™ DC persistent memory, supporting up-to 36 TB of system-level memory capacity when combine with traditional DRAM. 50 percent increased memory bandwidth and capacity. Support for six memory channels and up-to 4 TB of DDR4 memory, per socket, with speeds up-to 2933 MT/s (1 DPC)
  • Expanded I/O: 48 lanes of PCIe* 3.0 bandwidth and throughput for demanding I/O-intensive workloads
  • Intel® Ultra Path Interconnect (Intel® UPI): Four Intel® Ultra Path Interconnect (Intel® UPI) (9200 series) and up to three Intel® Ultra Path Interconnect (Intel® UPI) (8200 series) channels increase scalability of the platform to as many as two sockets (9200 series) and up to eight sockets (8200 series). Intel® Ultra Path Interconnect (Intel® UPI) offers the perfect balance between improved throughput and energy efficiency
  • Intel® Deep Learning Boost (Intel® DL Boost) with VNNI: New Intel® Deep Learning Boost (Intel® DL Boost) with Vector Neural Network Instruction (VNNI) bring enhanced artificial intelligence inference performance, with up to 30X performance improvement over the previous generation, 2nd Gen Intel® Xeon® Scalable processors help to deliver AI readiness across the data center, to the edge and back
  • Intel® Infrastructure Management Technologies (Intel® IMT): A framework for resource management, Intel® Infrastructure Management Technologies (Intel® IMT), combines multiple Intel capabilities that support platform-level detection, reporting, and configuration. This hardware-enhanced monitoring, management, and control of resources can help enable greater data center resource efficiency and utilization
  • Intel® Security Libraries◊ for Data Center (Intel® SecL-DC): A set of software libraries and components, Intel® SecL-DC enables Intel hardware-based security features. The open-source libraries are modular and have a consistent interface. They can be used by customers and software developers to more easily develop solutions that help secure platforms and help protect data using Intel hardware-enhanced security features at cloud scale
  • Intel® Advanced Vector Extensions 512 (Intel® AVX-512): With double the FLOPS per clock cycle compared to previous-generation Intel® Advanced Vector Extensions 2 (Intel® AVX2), 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
  • Security◊ without compromise: Limiting encryption overhead and performance on all secure data transactions

Innovative Integrations

Platform integrations deliver improvements in performance and latency across the infrastructure:

  • Integrated Intel® QuickAssist Technology (Intel® QAT): Chipset-based hardware acceleration for growing compression and cryptographic workloads for greater efficiency while delivering enhanced data transport and protection across server, storage, and network infrastructure
  • Integrated Intel® Ethernet with scalable iWARP* RDMA*: Provides up to four 10 Gbps high-speed Ethernet ports for high data throughput and low-latency workloads. Ideal for software-defined storage solutions, NVM Express* over Fabric solutions, and virtual machine migrations. Integrated in the chipset

Industry-leading Memory and Storage Support

Storage innovations can drive significant improvements in efficiency and performance of data-hungry workloads.

  • Support for Intel® Optane™ DC persistent memory: Breakthrough memory and storage memory innovation offering groundbreaking capabilities for fast storage solutions. Can be combined with Intel® Optane™ DC SSDs for the ultimate in storage and data performance
  • Support for Intel® Optane™ DC SSDs and Intel® QLC 3D NAND Solid-State Drives: Delivers industry-leading combination of high throughput, low latency, high QoS, and ultra-high endurance to break through data access bottlenecks
  • Deploy next-generation storage with confidence with Intel® Volume Management Device (Intel® VMD): Enables hot swapping of NVMe SSDs from the PCIe bus without shutting down the system, while standardized LED management helps provide quicker identification of SSD status. This commonality brings enterprise reliability, availability, and serviceability (RAS) features to NVMe SSDs, enabling deployment of next-generation storage with confidence
  • Intel® Intelligent Storage Acceleration Library (Intel® ISA-L): Optimizes storage operations, such as encryption, for increased storage performance

Complementary Offerings for Even Greater Performance, Scalability

Intel offers a broad hardware and software portfolio that complements this new processor.

  • Intel® Ethernet 800 series products support up to 100 GbE port speed with Application Device Queues (ADQ), which addresses latency-sensitive workloads for higher speed data communication. Data Plane Developer Kit (DPDK) is supported across Intel® Ethernet 800 series products for NFV acceleration, advanced packet forwarding, and highly efficient packet processing.
    Learn more at intel.com/ethernet
  • Intel® FPGAs offer flexible, programmable acceleration, for low-latency applications, such as virtual switching, network services, data analytics, and AI.
    Learn more at intel.com/fpga
  • A range of software tools and libraries for general and highly parallel computing help developers optimize applications for Intel® architecture.
    Learn more at software.intel.com

Enhanced Platform Trust
Data and platform reliability and protection are key concerns for enterprises dealing with increasing concerns and scrutiny regarding data security and privacy. Intel® Xeon® Scalable processor family helps build highly trusted infrastructures with platform data protection, resiliency, and uptime.

Increased Data Protection and Reliability Across Every Workload

  • Enhanced Intel® Run Sure Technology: New enhancements deliver advanced Reliability, Availability, and Serviceability (RAS) and server uptime for a company's most critical workloads. Hardware-assisted capabilities, including enhanced MCA and recovery and adaptive multidevice error correction, diagnose, and recover from previously fatal errors. And, they help ensure data integrity within the memory subsystem
  • Intel® Key Protection Technology (Intel® KPT) with Integrated Intel® QuickAssist Technology (Intel® QAT) and Intel® Platform Trust Technology (Intel® PTT): Deliver hardware-enhanced platform security◊ by providing efficient key and data protection at rest, in-use, and in-flight
  • Intel® Trusted Execution Technology (Intel® TXT) with One-Touch Activation: Enhanced platform security,◊ while providing simplified and scalable deployment for Intel® Trusted Execution Technology (Intel® TXT)

As more data-rich workloads flow through the data center, this comprehensive suite of hardware-enhanced features brings better data- and platform-level protection mechanisms for trusted services in enterprise and cloud environments.◊

Dynamic and Highly Efficient Service Delivery
The convergence of enhanced compute, memory, network, and storage performance, combined with software ecosystem optimizations, make Intel® Xeon® Scalable processor family the ideal platform for fully virtualized, software-defined data centers that dynamically self-provision resources—on-premises, through the network, and in the public cloud—based on workload needs.

Powerful Tools and Technologies for an Agile Data Center

Intel® Virtualization Technology (Intel® VT-x) features:

  • Mode-based execution control (MBE) virtualization: Provides an extra layer of protection from malware attacks in a virtualized environment by enabling hypervisors to more reliably verify and enforce the integrity of kernel level code
  • Timestamp counter scaling (TSC) virtualization: Provides workload optimization in hybrid cloud environments by allowing virtual machines to move across CPUs operating at different base frequencies

Intel® Node Manager (Intel® NM) 4.0: Helps IT intelligently manage and optimize power, cooling, and compute resources in the data center, maximizing efficiency, while reducing the chances of costly overheats.

Faster Time to Value with Intel® Select Solutions

In today's complex data center, hardware and software infrastructure is not “one size fits all". Intel® Select Solutions eliminates guesswork with rigorously benchmark tested and verified solutions optimized for real-world performance. These solutions accelerate infrastructure deployment on Intel® Xeon® processors for today's critical workloads in advanced analytics, hybrid cloud, storage, and networking.

Enterprise and Government – Primed for Business
For enterprise data centers modernizing to take advantage of the era of advanced analytics, the hybrid cloud and future-ready storage, Intel® Select Solutions can speed up your data-fueled, IT-driven business transformation.

Communications Service Providers – Tuned Network Enhancements
For Communication Service Providers transforming their network for a 5G-enabled future, Intel® Select Solutions offer a faster and more efficient deployment path of tested, reliable infrastructure with verified configurations that take full advantage of virtual network enhancements that support new and emerging customer workload demands.

High Performance Computing – Accelerated Time to Insight
For research in academia and government as well as the enterprises, high performance computing (HPC) capabilities with Intel® Select Solutions help push the limits of mainstream data today with deeper insights and more complex problem solving.

Learn more about Intel® Select Solutions featuring new 2nd Gen Intel® Xeon® Scalable processors with Intel® Optane™ DC persistent memory at intel.com/selectsolutions.

Strong, Capable Platforms for the Data-Fueled Enterprise

Enterprises are keen to extract value from the exploding data streams being presented to them for rapid insights that can shape their business initiatives. Traditional and emerging applications in the enterprise, including predictive analytics, machine learning, and HPC, require new levels of powerful compute capabilities and massive tiered data storage volumes. The modernized data center is being architected using a converged and holistic approach that can flexibly deliver new services and improve TCO across infrastructure assets today, while providing the most seamless and scalable on-ramp to a self-governing, hybrid data center.

Yet, organizations running their foundational business workloads, such as OLTP and web infrastructure, seek to reduce TCO with higher performing infrastructures.

The Intel® Xeon® Scalable processor family delivers next-generation enterprise capabilities to businesses through a future-ready platform that can serve the hybrid cloud, data-fueled era, plus it helps improve day-to-day operations. This versatile platform brings disruptive levels of compute performance, coupled with memory and I/O advances, to compute-hungry and latency-sensitive applications. Combined with innovative Intel® Optane™ DC persistent memory and Intel® QLC 3D NAND SSD data center family to manage large data volumes across storage, caching, and memory, platforms built on the Intel® Xeon® Scalable processor family are ready to handle the intense demands of the data and cloud era.

With a scalable portfolio of packaging options to suit diverse workload requirements, the Intel® Xeon® Scalable processor family is a performance workhorse designed for deploying highly efficient, virtualized infrastructures for compute, storage, and networking.

New 2nd Gen Intel® Xeon® Scalable Processors with Intel® Optane™ DC Persistent Memory

Highlights for Enterprise Innovation

  • 2nd Gen Intel® Xeon® Scalable processors
  • Intel® Optane™ DC persistent memory
  • Intel® Deep Learning Boost (Intel® DL Boost)
  • Intel® Speed Select Technology (Intel® SST)
  • Intel® Ethernet 800 Series
  • Intel® Optane™ DC SSDs and Intel® QLC 3D NAND SSDs
  • Intel® Infrastructure Management Technologies (Intel® IMT)

Next-generation Platform for Cloud-optimized, 5G-Ready Networks, and Next-generation Virtual Networks

The coming era of 5G will enable entirely new ecosystems and classes of consumer and enterprise services along with media applications on wireless and wireline networks. These data-rich, innovative use cases, driven by the new Internet of Things (IoT), visual computing, and analytics, represent significant future opportunities for communications service providers (CommSPs) to grow revenue.

The transition from purpose-built, fixed function infrastructure to a new generation of open networks is the essential first step to prepare for a 5G-enabled world. Software-defined networking with Network Functions Virtualization (NFV) is enabling new service opportunities and operations efficiencies for both communication services providers and enterprises alike. Using flexible, optimized, industry-standard servers and virtualized, orchestrated network functions will allow future-ready infrastructures to be able to deliver innovative services with efficiency and ease.

Such distributed communications networks can support extreme levels of scalability, agility, programmability, and security across an ever-growing volume and variety of networking workloads—from the network core to the edge.

The Intel® Xeon® Scalable processor family is the basis for next-generation platforms to build virtualized, cloud-optimized, 5G-ready networks. It offers an architecture that scales and adapts with ease to handle the demands of emerging applications and the convergence of key workloads, such as applications and services, control plane processing, high-performance packet processing, and signal processing. This new processor provides a foundation for agile networks that can operate with cloud economics, be highly automated and responsive, and support rapid and more secure delivery of new and enhanced services enabled by 5G.

New 2nd Gen Intel® Xeon® Scalable Processors Specialized for Networking/NFV “N” SKUs

Highlights for Communication Service Provider Innovation

  • 2nd Gen Intel® Xeon® Scalable processor “N” SKUs, specialized for Networking/NFV
  • Intel® Optane™ DC persistent memory
  • Hardware-based acceleration of encryption and compression using integrated Intel® QAT
  • Intel® Ethernet 800 Series
  • Intel® FPGAs maximize versatility in communications infrastructure
  • Intel® Infrastructure Management Technologies (Intel® IMT)

Additional Resources Optimized for Communication Service Providers
The open-source Data Plane Development Kit (DPDK) enables optimized communications operations on Intel® architecture. DPDK has demonstrated ability to scale performance as processor core count and performance increase; workloads, such as Vector Packet Processing (VPP) IPsec, benefit from this enhanced performance. Additionally, these libraries provide pre-optimized mechanisms to allow new processor capabilities (such as Intel® AVX-512 and memory and I/O enhancements) to be able to utilize the new functionality for improved packet processing performance with less direct development effort.

Intel offers programs, such as Intel® Network Builders University, ideal for network evolution in the 5G era. With solution guidance and training from these programs, CommSPs can drive their network transformation initiatives forward with increased confidence.

Breakthrough HPC and High-performance Data Analytics Innovation

Today's scientific discoveries are fueled by innovative algorithms, new sources and volumes of data, and advances in compute and storage. Benefitting from exponentially expanding volumes and variety of data, HPC clusters are also the engine for running evolving High-performance Data Analytics (HPDA) workloads, leading to incredible discoveries and insight for business and human understanding. Machine learning, deep learning, and AI converge the capabilities of massive compute with the flood of data to drive next-generation applications, such as autonomous systems and self-driving vehicles.

Intel® Xeon® Scalable processor family offers a common platform for AI with high throughput for both inference and training—up to 30x higher inference using 9200 series and up to 14x higher inference throughput using 8200 series, compared to Intel® Xeon® Scalable processors introduced in July 2017.

HPC is no longer just the domain of large scientific institutions. Enterprises are increasingly consuming a massive number of HPC compute cycles; some of the world's largest HPC clusters are in private oil and gas companies. Research in personalized medicine applies HPC for highly focused treatment plans. New HPC installations are engaging innovative, converged architectures for non-traditional usages that combine simulation, AI, visualization, and analytics in a single supercomputer.

HPC platforms—from the smallest clusters to largest supercomputers—demand a balance across compute, memory, storage, and network. The Intel® Xeon® Scalable processor family was designed to deliver and enable such balance with massive scalability—to tens of thousands of cores. From its improved core count and mesh architecture to newly integrated technologies and support for Intel® Optane™ DC persistent memory and storage devices, the Intel® Xeon® Scalable processor family enables the ultimate goals of HPC—to maximize performance across compute, memory, storage, and network without inducing bottlenecks at any intersection of resources.

The integration of Intel® Omni-Path Architecture (Intel® OPA), an end-to-end high-performance fabric, into the Intel® Xeon® Scalable processor family delivers both increased performance and scaling to distributed, parallel computing clusters. Near linear scaling up to 32 nodes enables building large HPC solutions that are not inhibited by the interconnect. The Intel® Xeon® Scalable processor family and Intel® Omni-Path Architecture (Intel® OPA) can enable new discoveries and faster solutions for highly parallel workloads in many data centers.

New 2nd Gen Intel® Xeon® Scalable Processors for HPC

  • Intel® Xeon® Platinum 9200 processors
  • Intel® Optane™ DC persistent memory
  • Intel® Ultra Path Interconnect (Intel® UPI)
  • Intel® Deep Learning Boost (Intel® DL Boost)
  • Intel® Advanced Vector Extensions 512 (Intel® AVX-512)
  • Intel® Omni-Path Architecture (Intel® OPA) Host Fabric Interface
  • Intel® Optane™ DC SSDs

Additional Technologies for HPC, HPDA, and AI
A range of high-productivity software tools, optimized libraries, foundational building blocks, and flexible frameworks for general and highly parallel computing help simplify workflows and assist developers to create codes that maximize the capabilities of IA for HPC and AI.

Optimizations for popular deep learning frameworks for IA, including Caffe and TensorFlow* offer increased value and performance for data scientists.

Intel® Parallel Studio XE 2017 includes performance libraries, such as Intel® Math Kernel Library (Intel® MKL) for Deep Neural Networks (Intel® MKL-DNN) to accelerate deep learning frameworks on IA, and Intel® Data Analytics Acceleration Library (Intel® DAAL) to speed big data analytics.

Resources Optimized for HPC
To continue to advance discovery through HPCinto the Exascale era, the Intel® Modern Code Developer Community offers developers and data scientists easily accessible online and face-to-face code modernization technical sessions on techniques, such as vectorization, memory and data layout, multithreading, and multinode programming.

Overview of 2nd Gen Intel® Xeon® Scalable Processors

Intel® Xeon® Scalable processor family

Intel® Xeon® Platinum 9200 Processors

Designed for high performance computing, advanced artificial intelligence and analytics, the Intel® Xeon® Platinum 9200 processors deliver breakthrough levels of performance with the highest Intel® architecture FLOPS per rack, along with the highest DDR native memory bandwidth support of any Intel® Xeon® processor platform.

  • Up to 56 Intel® Xeon® Scalable processing cores per processor
    Two processors per 2U platform
    (Intel® Server System S9200WK Data Center Block)
  • 12 memory channels per processors, 24 memory channels per node
  • Features new Intel® Deep Learning Boost (Intel® DL Boost) instruction for enhanced AI inference acceleration and performance
  • Enhanced multichip package optimized for density and performance

Intel® Xeon® Platinum 8200 Processors

Second-Generation Intel® Xeon® Platinum processors are the foundation for secure, agile, hybrid cloud data centers. With enhanced hardware-based security and exceptional two, four, and eight+ socket processing performance, these processors are built for mission Critical, real-time analytics, machine learning, artificial intelligence and multicloud workloads. With trusted, hardware-enhanced data service delivery, this processor family delivers monumental leaps in I/O, memory, storage, and network technologies to harness actionable insights from our increasingly data-fueled world.

Intel® Xeon® Gold 6200 Processors and Intel® Xeon® Gold 5200 Processors

With support for the higher memory speeds, enhanced memory capacity, and four-socket scalability, the Intel® Xeon® Gold 6200 processors deliver significant improvement in performance, advanced reliability, and hardware-enhanced security. It is optimized for demanding mainstream data center, multicloud compute, and network and storage workloads. The Intel® Xeon® Gold 5200 processors deliver improved performance with affordable advanced reliability and hardware-enhanced security. With up-to four-socket scalability, it is suitable for an expanded range of workloads.

Intel® Xeon® Silver 4200 Processors

Intel® Xeon® Silver processors deliver essential performance, improved memory speed, and power efficiency. Hardware-enhanced performance required for entry data center computes, network, and storage.

Entry-level Performance and HW-enhanced Security

The Intel® Xeon® Bronze processors delivers entry performance for small business and basic storage servers. Hardware-enhanced reliability, availability, and serviceability features designed to meet the needs of these entry solutions.

 

Intel® Xeon® Bronze Processor (3200 Series)

Intel® Xeon® Silver Processor (4200 Series)

 Intel® Xeon® Gold Processor (5200 Series)

Intel® Xeon®

Gold Processor

(6200 Series)

Intel® Xeon® Platinum Processor

(8200 Series)

Intel® Xeon® Platinum Processor
(9200 Series)
Pervasive Performance and Security

Highest Core Count Supported

6 cores

16 cores

18 cores

24 cores

28 cores

56 cores

Highest Supported Frequency

1.9 GHz

3.5 GHz

3.9 GHz

4.4 GHz

4.0 GHz

3.8 GHz

Number of CPU Sockets Supported

Up to 2

Up to 2

Up to 4

Up to 4

Up to 8

Up to 2

Intel® Ultra Path Interconnect (Intel® UPI)

2

2

2

3

3

4

Intel® UPI Speed

9.6 GT/s

9.6 GT/s

10.4 GT/s

10.4 GT/s

10.4 GT/s

10.4 GT/s

Intel® Advanced Vector Extensions 512 (Intel® AVX-512)

1 FMA

1 FMA

1 FMA

2 FMA

2 FMA

2 FMA

Memory Speed Support (DDR4)

2133 MT/s

2400 MT/s

2666 MT/s

2933 MT/s

2933 MT/s

2933 MT/s

Highest Memory Capacity Supported Per Socket◊

1 TB

1 TB

1 TB, 2 TB, 4.5 TB

1 TB, 2 TB, 4.5 TB

1 TB, 2 TB, 4.5 TB

1.5 TB

16Gb DDR4 DIMM Support

*

*

 *

*

*

*

Intel® Deep Learning Boost (Intel® DL Boost) with Vector Neural Network Instruction (VNNI)

*

*

*

*

*

*

Intel® Optane™ DC Persistent Memory Module Support◊

 

*

*

*

*

 

Intel® Omni-Path Architecture (Discrete PCIe* card)

*

*

*

*

*

*

Intel® QuickAssist Technology (Integrated in chipset)

*

*

*

*

*

 

Intel® QuickAssist Technology (Discrete PCIe card)

*

*

*

*

*

*

Intel® Optane™ DC SSDs

*

*

*

*

*

*

Intel® DC SSD Data Center Family (3D NAND)

*

*

*

*

*

*

PCIe 3.0

*

*

*

*

*

*

Intel® QuickData Technology (CBDMA)

*

*

*

*

*

*

Non-Transparent Bridge (NTB)

*

*

*

*

*

*

Intel® Turbo Boost Technology 2.0

 

*

*

*

*

*

Intel® Hyper-Threading Technology (Intel® HT Technology)

 

*

*

*

*

*

Node Controller Support

 

 

 

*

*

*
◊ Supported on select processors only
High Reliability

Reliability, Availability, and Serviceability (RAS) Capability

Standard

Standard

Standard

Advanced

Advanced

Advanced

Intel® Run Sure Technology

 

 

 

*

*

*
Agility & Efficiency

Intel® Speed Select Technology (Intel® SST)◊

 

*

*

*

*

 

Intel® Infrastructure Management Technologies

*

*

*

*

*

*

Intel® Resource Director Technology (Intel® RDT)

*

*

*

*

*

*

Intel® Volume Management Device (Intel® VMD)

*

*

*

*

*

*

Intel® Virtualization Technology (Intel® VT)

*

*

*

*

*

*

Intel® Speed Shift Technology

*

*

*

*

*

*

Intel® Node Manager 4.0

*

*

*

*

*

*

Mode-based Execute Control

*

*

*

*

*

*

Timestamp Counter Scaling (TSC) for Virtualization

*

*

*

*

*

*
◊ Supported on select processors only
Security

Intel® Security Libraries for Data Center (Intel® ISecL-DC)

*

*

*

*

*

*

Intel® Advanced Vector Extensions 512 (Intel® AVX-512)

*

*

*

*

*

*

Intel® Key Protection Technology (KPT) w/Integrated Intel® QAT

*

*

*

*

*

*

Intel® Platform Trust Technology (PTT)

*

*

*

*

*

*

Intel® Trusted Execution Technology (Intel® TXT) w/ One-Touch Activation (OTA)

*

*

*

*

*

*
Voir plus Voir moins View all

2nd Gen Intel® Xeon® Scalable Processors

SKU Numbering
Processor numbers for the Intel® Xeon® Scalable processor family use an alphanumeric scheme based on performance, features, processor generation, and any options, following the brand and its class.

Profitez davantage de votre investissement dans le Cloud

Atteignez le considérable rendement par dollar dont vous avez besoin avec la technologie Intel®. Les charges de travail critiques et gourmandes en données, telles que les bases de données, le calcul intensif (HPC) et le Web, deviennent plus efficaces pour un moindre coût total de possession avec les Clouds basés sur l'architecture Intel®.3 4 5 6 7

Performances 2,84 fois plus élevées

Un rapport performances/prix jusqu'à 2,84 fois plus élevé pour les charges de travail de bases de données, y compris HammerDB PostgreSQL* et MongoDB*.3

En savoir plus

Performances 4,15 fois plus élevées

Un rapport performances/prix jusqu'à 4,15 fois plus élevé sur High Performance LINPACK* et LAMMPS*.4

En savoir plus

Performances 1,74 fois plus élevées

Un rapport performances/prix plus élevé sur Java côté serveur et un rapport performances/prix 1,74 fois plus élevé sur Wordpress PHP/HHVM.5

En savoir plus

Performances 2,25 fois plus élevées

Un rapport performances/prix jusqu'à 2,25 fois plus élevé pour les applications de bande passante mémoire.6

En savoir plus

Voir plus Voir moins

2nd Gen Intel® Xeon® Scalable Processor SKUs

For the most up-to-date information, please visit intel.com/xeon or ark.intel.com

Processor Identifier

Core Count

Cache (MB)

TDP (WATTS)

Processor Base Frequency (GHz)

Max Turbo Frequency Rate (GHz)

Max Memory Speed (MT/s)

Memory Capacities
M/L SKU Suffix

PLATINUM 9282

56

77

400

2.6

3.8

2933

3TB

PLATINUM 9242

48

71.5

350

2.3

3.8

2933

3TB

PLATINUM 9222◊

32

71.5

250

2.3

3.7

2933

3TB

PLATINUM 8280

28

38.5

205

2.7

4

2933

1TB / 2TB / 4.5TB

PLATINUM 8276

28

38.5

165

2.2

4

2933

1TB / 2TB / 4.5TB

PLATINUM 8270

26

35.75

205

2.7

4

2933

1TB

PLATINUM 8268

24

35.75

205

2.9

3.9

2933

1TB

PLATINUM 8260

24

35.75

165

2.4

3.9

2933

1TB / 2TB / 4.5TB

PLATINUM 8256

4

16.5

105

3.8

3.9

2933

1TB

PLATINUM 8253

16

22

125

2.2

3

2933

1TB

GOLD 6254

18

24.75

200

3.1

4

2933

1TB

GOLD 6252

24

35.75

150

2.1

3.7

2933

1TB

GOLD 6248

20

27.5

150

2.5

3.9

2933

1TB

GOLD 6246

12

24.75

165

3.3

4.2

2933

1TB

GOLD 6244

8

24.75

150

3.6

4.4

2933

1TB

GOLD 6242

16

22

150

2.8

3.9

2933

1TB

GOLD 6240

18

24.75

150

2.6

3.9

2933

1TB / 2TB / 4.5TB

GOLD 6238

22

30.25

140

2.1

3.7

2933

1TB / 2TB / 4.5TB

GOLD 6234

8

24.75

130

3.3

4

2933

1TB

GOLD 6230

20

27.5

125

2.1

3.9

2933

1TB

GOLD 6226

12

19.25

125

2.7

3.7

2933

1TB

GOLD 5222

4

16.5

105

3.8

3.9

2933

1TB

GOLD 5220

18

24.75

125

2.2

3.9

2667

1TB

GOLD 5218

16

22

125

2.3

3.9

2667

1TB

GOLD 5217

8

11

115

3

3.7

2667

1TB

GOLD 5215

10

13.75

85

2.5

3.4

2667

1TB / 2TB / 4.5TB

SILVER 4216

16

22

100

2.1

3.2

2400

1TB

SILVER 4215

8

11

85

2.5

3.5

2400

1TB

SILVER 4214

12

16.5

85

2.2

3.2

2400

1TB

SILVER 4210

10

13.75

85

2.2

3.2

2400

1TB

SILVER 4208

8

11

85

2.1

3.2

2400

1TB

BRONZE 3204

6

8.25

85

1.9

1.9

2133

1TB

Networking/NFV Specialized Featuring Intel® Speed Select Technology (Intel® SST) SKUs

6252N

24

35.75

150

2.3

3.6

2933

1TB

6230N

20

27.5

125

2.3

3.9

2933

1TB

5218N

16

22

105

2.3

3.7

2667

1TB

Search Application Value Specialized SKU

5220S

18

24.75

125

2.7

3.9

2667

1TB

Long-Life Cycle and NEBS-Thermal Friendly SKUs

6238T

22

30.25

125

1.9

3.7

2933

1TB

6230T

20

27.5

125

2.1

3.9

2933

1TB

5220T

18

24.75

105

1.9

3.9

2667

1TB

5218T

16

22

105

2.1

3.8

2667

1TB

4209T

8

11

70

2.2

3.2

2400

1TB

VM Density Value Specialized SKUs

6262V

24

33

135

1.9

3.6

2400

1TB

6222V

20

27.5

115

1.8

3.6

2400

1TB

SKUS Featuring Intel® Speed Select Technology (Intel® SST)

8260Y

24

35.75

165

2.4

3.9

2933

1TB

6240Y

18

24.75

150

2.6

3.9

2933

1TB

4214Y

12

16.5

85

2.2

3.2

2400

1TB
◊ Intel® Xeon® Platinum 9222 processor: Intel® Turbo Boost Technology frequency and base frequency are preliminary; final values to be published at a future date.
Voir plus Voir moins View all

PRODUCT NAME

SKU

10Gb/1Gb ETHERNET PORTS

COMPRESSION

ENCRYPTION

RSA

 

Intel® QuickAssist Technology (Intel® QAT)

Intel® C621 chipset

LBG-1G

0/4

N/A

N/A

N/A

Intel® C622 chipset

LBG-2

2/4

N/A

N/A

N/A

Intel® C624 chipset

LBG-4

4/4

N/A

N/A

N/A

Intel® C625 chipset

LBG-E

4/4

20 Gb/s

20 Gb/s

20K Ops/s

Intel® C626 chipset

LBG-M

4/4

40 Gb/s

40 Gb/s

40K Op/s

Intel® C627 chipset

LBG-T

4/4

100 Gb/s

100 Gb/s

100K Ops/s

Intel® C628 chipset

LBG-L

4/4

100 Gb/s

100 Gb/s

100K Ops/s
Voir plus Voir moins View all

Intel® Server System S9200WK Data Center Block (DCB) Technical Specifications
supporting the Intel® Xeon® Platinum 9200 Processor

Form Factor

2U rack enclosure; Up to 4 independent warm-swap compute modules

CPU

Intel® Xeon® Platinum 9200 processors up to 56 cores

Memory

DDR4-2933 MT/s DIMMS, 24 DIMMs per platform @ 1DPC

Supports 8GB to 128GB DIMM options, number, and capacity configurable

Storage

Up to 8x M.2 SSDs per DCB (2x per compute module with 1U compute module; Up to 4x M.2 SSDs & 4x hot-swap U.2 NVMe SSDs with 2U Compute Modules)

M.2 and U.2 number and capacity configurable

Ethernet

Integrated 1Gbase-T RJ45 (two ports per compute module), Optional shared 1Gbase-T RJ45 management port chassis card

Cooling

Available with high flow rate air-cooling or integrated liquid-cooling options

I/O

2 x16 Gen3 PCIe slots per 1U compute module; 4 x16 Gen3 PCIe slots per 2U compute module for high-speed networking support

Manageability

Dedicated, consolidated Management Module

Security & Serviceability

TPM 2.0 (optional); Hot swap/redundant fans, and PSUs; light path diagnostic LEDs
Not a comprehensive list of all features and capabilities. Learn more at intel.com/serverproducts
Voir plus Voir moins View all

Download the brief for a full list of features ›

2nd Gen Intel® Xeon® Scalable Processors


More Information

Intel® technologies’ features and benefits depend on system configuration and may require enabled hardware, software, or service activation. Performance varies depending on system configuration.

◊ No product or component can be absolutely secure.

Tests document performance of components on a particular test, in specific systems. Differences in hardware, software, or configuration will affect actual performance. For more complete information about performance and benchmark results, visit https://www.intel.com/benchmarks.

Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations, and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit https://www.intel.com/benchmarks.

Intel® Advanced Vector Extensions (Intel® AVX) provides higher throughput to certain processor operations. Due to varying processor power characteristics, utilizing Intel® Advanced Vector Extensions (Intel® AVX) instructions may cause a) some parts to operate at less than the rated frequency and b) some parts with Intel® Turbo Boost Technology 2.0 to not achieve any or maximum turbo frequencies. Performance varies depending on hardware, software, and system configuration and you can learn more at https://www.intel.com/go/turbo.

Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Cost reduction scenarios described are intended as examples of how a given Intel-based product, in the specified circumstances and configurations, may affect future costs and provide cost savings. Circumstances will vary. Intel does not guarantee any costs or cost reduction.

Intel does not control or audit third-party benchmark data or the web sites referenced in this document. You should visit the referenced web site and confirm whether referenced data are accurate.

© Intel Corporation. Intel, the Intel logo, Intel Optane DC, and Intel Xeon are trademarks of Intel Corporation in the U.S. and/or other countries.
* Other names and brands may be claimed as property of others.

FTC Optimization Notice
If you make any claim about the performance benefits that can be achieved by using Intel® Software Development Tools (compilers or libraries), you must use the entire text on the same viewing plane (slide) as the claim.

Optimization Notice: Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice Revision #20110804
Performance results are based on testing as of dates shown in configuration and may not reflect all publicly available security updates. See configuration disclosure for details. No product or component can be absolutely secure. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations, and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit www.intel.com/benchmarks.

Pricing guidance as of March 1, 2019. Intel does not guarantee any costs or cost reduction. You should consult other information and performance tests to assist you in your purchase decision.

Infos sur le produit et ses performances

1

Source: IDC, http://www.idc.com/getdoc.jsp?containerId=prUS41888916

2

Source: Gartner Group, http://www.gartner.com/newsroom/id/3142917

3

Résultats calculés par Intel P2CA à l'aide de tarifs AWS ($/heure, durée standard de 1 an, aucun investissement initial) en date du 12 janvier 2019.
Tests de performances/prix réalisés sur des instances AWS* EC2 R5 et R5a (https://aws.amazon.com/ec2/instance-types/), comparant les performances/prix d'un processeur Intel® Xeon® Scalable 96 vCPU aux performances/prix d'un processeur AMD EPYC*.

Charge de travail : HammerDB* PostgreSQL*
Résultats : performances/prix AMD EPYC = ligne de référence de 1 ; performances/prix du processeur Intel® Xeon® Scalable = 1,85X (le score le plus élevé étant le meilleur)
Base de données : HammerDB - PostgreSQL (le score le plus élevé étant le meilleur) :
Instance (Intel) AWS R5.24xlarge, HammerDB 3.0 PostgreSQL 10.2, mémoire : 768 Go, hyperviseur : KVM ; type de stockage : EBS io1, volume de disque 200 Go, 200 Go de stockage total, version Docker : 18.06.1-ce, RedHat* Enterprise Linux 7.6, 3.10.0-957.el7.x86_64, 6 400 Mo de mémoire tampon partagée, 256 entrepôts, 96 utilisateurs. Score « NOPM » 439 931, mesure réalisée par Intel les 11/12/18 et 14/12/18.
Instance (AMD) AWS R5a.24xlarge, HammerDB 3.0 PostgreSQL 10.2, mémoire : 768 Go, hyperviseur : KVM ; type de stockage : EBS io1, volume de disque 200 Go, 200 Go de stockage total, version Docker : 18.06.1-ce, RedHat* Enterprise Linux 7.6, 3.10.0-957.el7.x86_64, 6 400 Mo de mémoire tampon partagée, 256 entrepôts, 96 utilisateurs. Score « NOPM » 212 903, mesure réalisée par Intel le 20/12/2018.

Charge de travail : MongoDB*
Résultats : performances/prix AMD EPYC = ligne de référence de 1 ; performances/prix du processeur Intel® Xeon® Scalable = 2,84X (le score le plus élevé étant le meilleur)
Base de données : MongoDB (le score le plus élevé étant le meilleur) :
Instance (Intel) AWS R5.24xlarge, MongoDB v4.0, journal désactivé, synchronisation avec système de fichiers désactivée, cache WiredTiger = 27 Go, maxPoolSize = 256 ; 7 instances MongoDB, 14 machines virtuelles client, 1 client YCSB par machine virtuelle, 96 threads par client YCSB, RedHat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, score 1 229 288 ops/sec, mesure réalisée par Intel le 10/12/18.
Instance (AMD) AWS R5a.24xlarge, MongoDB v4.0, journal désactivé, synchronisation avec système de fichiers désactivée, cache WiredTiger = 27 Go, maxPoolSize = 256 ; 7 instances MongoDB, 14 machines virtuelles client, 1 client YCSB par machine virtuelle, 96 threads par client YCSB, RedHat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, score 388 596 ops/sec, mesure réalisée par Intel le 10/12/18.
Pour en savoir plus, consultez www.intel.fr/benchmarks.

4

Résultats calculés par Intel P2CA à l'aide de tarifs AWS ($/heure, durée standard de 1 an, aucun investissement initial) en date du 12 janvier 2019.
Tests de performances/prix réalisés sur des instances AWS* EC2 M5 et M5a (https://aws.amazon.com/ec2/instance-types/), comparant les performances/prix d'un processeur Intel® Xeon® Scalable 96 vCPU aux performances/prix d'un processeur AMD EPYC*.

Charge de travail : LAMMPS*
Résultats : performances/prix AMD EPYC = ligne de référence de 1 ; performances/prix du processeur Intel® Xeon® Scalable = 4,15X (le score le plus élevé étant le meilleur)
Science des matériaux HPC : LAMMPS (le score le plus élevé étant le meilleur) :
Instance (Intel) AWS M5.24xlarge, version LAMMPS : 22/08/2018 (Code : https://lammps.sandia.gov/download.html), charge de travail : eau - 512 000 particules, Intel ICC 18.0.3.20180410, bibliothèque MPI Intel® pour système d'exploitation Linux*, version 2018 mise à jour 3 Build 20180411, 48 rangs MPI, RedHat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, OMP_NUM_THREADS=2, score 137,5 pas de temps/sec, mesure réalisée par Intel le 31/10/18.
Instance (AMD) AWS M5a.24xlarge, version LAMMPS : 22/08/2018 (Code : https://lammps.sandia.gov/download.html), charge de travail : eau - 512 000 particules, Intel ICC 18.0.3.20180410, bibliothèque MPI Intel® pour système d'exploitation Linux*, version 2018 mise à jour 3 Build 20180411, 48 rangs MPI, RedHat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, OMP_NUM_THREADS=2, score 55,8 pas de temps/sec, mesure réalisée par Intel le 07/11/2018.
Modifications AMD pour prendre en charge AVX2 (AMD ne prenant en charge qu'AVX2, ces modifications étaient nécessaires) :
sed -i 's/-xHost/-xCORE-AVX2/g' Makefile.intel_cpu_intelmpi
sed -i 's/-qopt-zmm-usage=high/-xCORE-AVX2/g' Makefile.intel_cpu_intelmpi

Charge de travail : High Performance Linpack*
Résultats : performances/prix AMD EPYC = ligne de référence de 1 ; performances/prix du processeur Intel® Xeon® Scalable = 4,15X (le score le plus élevé étant le meilleur)
Linpack HPC (le score le plus élevé étant le meilleur) :
Instance (Intel) AWS M5.24xlarge, version HP Linpack 2.2 (https://software.intel.fr/en-us/articles/intel-mkl-benchmarks-suite Répertoire : benchmarks_2018.3.222/linux/mkl/benchmarks/mp_linpack/bin_intel/intel64), Intel ICC 18.0.3.20180410 avec AVX512, bibliothèque MPI Intel® pour système d'exploitation Linux*, version 2018 mise à jour 3 Build 20180411, RedHat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, OMP_NUM_THREADS=24, 2 processus MPI, score 3 152 Gbit/s, mesure réalisée par Intel le 31/10/18.
Instance (AMD) AWS M5a.24xlarge, version HP Linpack 2.2, (source HPL : http://www.netlib.org/benchmark/hpl/hpl-2.2.tar.gz ; version 2.2 ; icc (ICC) 18.0.2 20180210 ; compilait et établissait un lien vers la bibliothèque BLIS version 0.4.0 ; https://github.com/flame/blis ; drapeaux Addt'l Compiler : -O3 -funroll-loops -W -Wall –qopenmp ; make arch=zen OMP_NUM_THREADS=8 ; 6 processus MPI). Intel ICC 18.0.3.20180410 avec AVX2, bibliothèque MPI Intel® pour le système d'exploitation Linux*, version 2018 mise à jour 3 Build 20180411, RedHat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, OMP_NUM_THREADS=8, 6 processus MPI, score 677,7 Gbit/s, mesure réalisée par Intel le 07/11/18.

5

Résultats calculés par Intel P2CA à l'aide de tarifs AWS ($/heure, durée standard de 1 an, aucun investissement initial) en date du 12 janvier 2019.
Tests de performances/prix réalisés sur des instances AWS* EC2 M5 et M5a (https://aws.amazon.com/ec2/instance-types/), comparant les performances/prix d'un processeur Intel® Xeon® Scalable 96 vCPU aux performances/prix d'un processeur AMD EPYC*.

Charge de travail : Java côté serveur* 1 JVM
Résultats : performances/prix AMD EPYC = ligne de référence de 1 ; performances/prix du processeur Intel® Xeon® Scalable = 1,74X (le score le plus élevé étant le meilleur)
Java côté serveur (le score le plus élevé étant le meilleur) :
Instance (Intel) AWS M5.24xlarge, banc d'essai serveur Java sans liaison NUMA, 2JVM, OpenJDK 10.0.1, RedHat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, score 101 767 transactions/sec, mesure réalisée par Intel le 16/11/18.
Instance (AMD) AWS M5a.24xlarge, banc d'essai serveur Java sans liaison NUMA, 2JVM, OpenJDK 10.0.1, RedHat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, score 52 068 transactions/sec, mesure réalisée par Intel le 16/11/18.

Charge de travail : Wordpress* PHP/HHVM*
Résultats : performances/prix AMD EPYC = ligne de référence de 1 ; performances/prix du processeur Intel® Xeon® Scalable = 1,75X (le score le plus élevé étant le meilleur)
Wordpress Web front-end (le score le plus élevé étant le meilleur) :
Instance (Intel) AWS M5.24xlarge, oss-performance/wordpress Ver 4.2.0 ; Ver 10.2.19-MariaDB-1:10.2.19+maria~bionic ; version charge de travail : u'4.2.0 ; threads client : 200 ; PHP 7.2.12-1 ; perfkitbenchmarker_version="v1.12.0-944-g82392cc ; Ubuntu 18.04, noyau Linux 4.15.0-1025-aws, score 3 626,11 TPS, mesure réalisée par Intel le 16/11/18.
Instance (AMD) AWS M5a.24xlarge, oss-performance/wordpress Ver 4.2.0 ; Ver 10.2.19-MariaDB-1:10.2.19+maria~bionic ; version charge de travail : u'4.2.0 ; threads client : 200 ; PHP 7.2.12-1 ; perfkitbenchmarker_version="v1.12.0-944-g82392cc ; Ubuntu 18.04, noyau Linux 4.15.0-1025-aws, score 1 838,48 TPS, mesure réalisée par Intel le 16/11/18.
Pour en savoir plus, consultez www.intel.fr/benchmarks.

6

Instance (Intel) AWS M5.4xlarge, Flux McCalpin (version OMP), (Source :https://www.cs.virginia.edu/stream/FTP/Code/stream.c) ; Intel ICC 18.0.3.20180410 avec AVX512, -qopt-zmm-usage=high, -DSTREAM_ARRAY_SIZE=134217728 -DNTIMES=100 -DOFFSET=0 –qopenmp, -qoptstreaming-stores always -o $OUT stream.c, Red Hat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, OMP_NUM_THREADS: 8, KMP_AFFINITY: proclist=[0-7:1], granularity=thread, explicit, score 81216.7 Mo/s, mesure réalisée par Intel le 06/12/18.
Instance (AMS) AWS M5a.4xlarge, Flux McCalpin (version OMP), (Source :https://www.cs.virginia.edu/stream/FTP/Code/stream.c) ; Intel ICC 18.0.3.20180410 avec AVX2,-DSTREAM_ARRAY_SIZE=134217728, -DNTIMES=100 -DOFFSET=0 -qopenmp -qopt-streaming-stores always -o $OUT stream.c, Red Hat* Enterprise Linux 7.5, noyau 3.10.0-862.el7.x86_64, OMP_NUM_THREADS: 8, KMP_AFFINITY : proclist=[0-7:1], granularity=thread,explicit, score 32154.4 Mo/s, mesure réalisée par Intel le 06/12/18.
Avertissement OpenFOAM : cette offre n'est pas approuvée ni avalisée par OpenCFD Limited, producteur et distributeur du logiciel OpenFOAM via www.openfoam.com, et propriétaire des marques commerciales OpenFOOAM* et OpenCFD*.
Tarifs AWS en date du 12 janvier 2019 : tarification des instances réservées pour une durée standard de 1 an (https://aws.amazon.com/ec2/pricing/reserved-instances/pricing/) Tarifs à la demande par heure Linux/Unix (https://aws.amazon.com/ec2/pricing/on-demand/).

7

Amélioration multipliée par 30 du débit d'inférence sur le processeur Intel® Xeon® Platinum 9282 avec la technologie Intel® Deep Learning Boost (Intel® DL Boost) : test réalisé par Intel le 26 février 2019. Plate-forme : processeur Intel® Xeon® Platinum 9282 2 sockets Dragon rock (56 cœurs par socket), technologie Hyper Threading ACTIVÉE, technologie Turbo ACTIVÉE, 768 Go de mémoire totale (24 emplacements / 32 Go / 2 933 MHz), BIOS : SE5C620.86B.0D.01.0241.112020180249, noyau CentOS 7 3.10.0-957.5.1.el7.x86_64, structure de Deep Learning : optimisation Intel® pour version Caffe* : https://github.com/intel/caffe d554cbf1, ICC 2019.2.187, MKL DNN version : v0.17 (hachage de validation : 830a10059a018cd2634d94195140cf2d8790a75a), modèle : https://github.com/intel/caffe/blob/master/models/intel_optimized_models/int8/resnet50_int8_full_conv.prototxt, BS=64, Aucune couche de données syntheticData : 3x224x224, 56 instances/2 sockets, type de données : INT8 ; vs. Testé par Intel en date du 11 juillet 2017 : processeur Intel® Xeon® Platinum 8180 2 sockets à 2,50 GHz (28 cœurs), technologie Hyper Threading désactivée, technologie turbo désactivée, mode d'échelonnage réglé sur « performances » via le pilote intel_pstate, RAM ECC DDR4-2666 384 Go. CentOS Linux* version 7.3.1611 (Core), noyau Linux* 3.10.0-514.10.2.el7.x86_64. Unité de stockage SSD Intel® série S3700 pour les centres de données (800 Go, 2,5 pouces, SATA 6 Gbit/s, 25 nm, MLC). Performances mesurées avec : variables environnementales : KMP_AFFINITY='granularity=fine, compact‘, OMP_NUM_THREADS=56, fréquence du processeur réglée avec les performances CPU Power frequency-set-d 2,5G -u 3,8G -g. Caffe : (http://github.com/intel/caffe/), révision f96b759f71b2281835f690af267158b82b150b5c. Inférence mesurée avec la commande « caffe time --forward_only », formation mesurée avec la commande « caffe time ». Un ensemble de données synthétique a été utilisé pour les topologies « ConvNet ». Pour les autres topologies, les données ont été stockées localement et mises en mémoire avant la formation. Spécifications des topologies sur https://github.com/intel/caffe/tree/master/models/intel_optimized_models(ResNet-50). Compilateur C++ Intel® ver. 17.0.2 20170213, Intel® Math Kernel Library (Intel® MKL) small libraries version 2018.0.20170425. Caffe exécuté avec « numactl -l ».