In one of the most important product launches in its history, Intel introduced its new products that provide a huge increase in data center performance, efficiency, security and new capabilities for artificial intelligence, cloud, network, edge and the world’s most powerful supercomputers.
Promoted products; There were 4th generation Intel Xeon scalable and Intel Xeon Max series processors codenamed Sapphire Rapids, as well as Intel Max series GPUs codenamed Ponte Vecchio.
Intel’s unique approach to providing purpose-built, workload-first, and highly optimized software tuned for specific workloads ensures that the company delivers the right performance at the right power at an optimal total cost of ownership.
4th Generation Xeon Scalable Processors
As Intel’s most sustainable data center processors, 4th Gen Intel Xeon Scalable processors provide customers with a set of features to manage power and performance, ensuring optimal use of CPU resources to help them meet their sustainability goals.
Sandra Rivera, Intel Vice President and General Manager, Data Center and Artificial Intelligence Group;
“4. The launch of the 2nd Generation Xeon Scalable processors and Max Series product family marks a pivotal moment in Intel’s acceleration of turnaround, reopening our path to data center leadership, and growing our footprint in new areas.
Intel’s 4th Gen Xeon and Max Series product family fulfills what customers want—leading performance and reliability in a secure environment for real-world needs—and accelerates innovation speeds by reducing time to value.” said.
The 4th Gen Xeon family significantly expands Intel’s purposeful and workload-first strategy and approach, unlike any other data center processor on the market and in the hands of customers today.
Leading Performance and Sustainability Benefits with Most Built-in Acceleration
There are over 100 million Xeons installed on the market today. From on-premises servers running IT services, including new “as-a-service” business models, to network equipment that manages internet traffic, to wireless base station computing at the edge, and cloud services.
Building on decades of data center, network, and smart edge innovation and leadership, new 4th Gen Xeon processors are designed to enable customers to tackle their most important computing challenges in AI, analytics, networking, security, storage, and HPC. It provides leading performance with most built-in accelerators.
When using built-in accelerators, 4th Gen Xeon customers benefit from an average of 2.9x performance efficiency gains per watt for targeted workloads, up to 70-watt power savings in optimized power mode with minimal performance loss, and 52 percent to 66 percent lower total cost of ownership when using built-in accelerators compared to previous generations. (TOC) can wait to benefit.
Sustainability
The broad scope of built-in accelerators in 4th Gen Xeon enables Intel to deliver platform-level power savings, reducing the need for additional discrete acceleration and helping our customers meet their sustainability goals.
In addition, the new Optimized Power Mode can deliver up to 20 percent power savings with less than 5 percent performance impact for select workloads. New innovations in air and liquid cooling are further reducing overall data center energy consumption. Intel plants used at least 90 percent renewable energy with state-of-the-art water treatment to produce the 4th Generation Xeon.
Artificial intelligence
Compared to the previous generation, 4th Gen Xeon processors achieve up to 10x higher PyTorch real-time inference and training performance in artificial intelligence with built-in Intel Advanced Matrix Extension (Intel AMX) accelerators.
Intel’s 4th Gen Xeon enables new levels of performance for inference and training across a wide array of AI workloads. The Xeon CPU Max Series, on the other hand, extends these capabilities for natural language processing, enabling customers to achieve up to 20x speed gains on major language models.
Intel’s AI software suite gives developers the ability to use their preferred AI tool, increasing productivity and accelerating AI development time. Portable from the workstation, this package can scale anywhere from the cloud to the edge.
And this has been confirmed by more than 400 machine learning and deep learning AI models in the most common AI use cases in every business segment.
Networking
4th Generation Xeon; It offers a family of processors specifically optimized for high-performance, low-latency networking and edge workloads. These processors are a critical component of the foundation leading to a more software-defined future for many industries, from telecommunications and retail to manufacturing and smart cities.
For 5G core workloads, built-in accelerators help increase throughput and reduce latency, while advances in power management improve both the responsiveness and efficiency of the platform. And the 4th Gen Xeon offers up to twice the virtualized radio access network (vRAN) capacity of previous generations without increasing power consumption. Thus, communication service providers; It can double the performance per watt to meet critical performance, scale and energy efficiency needs.
High Performance Computing
The 4th Gen Xeon and Intel Max Series product family enables solving the world’s toughest problems with a scalable and balanced architecture that integrates CPU and GPU into oneAPI’s open software ecosystem for demanding computing workloads in HPC and AI.
The first and only x86-based processor with high-bandwidth memory, the Xeon CPU Max Series accelerates many HPC workloads without the need to make code changes. The Intel Data Center GPU Max Series, Intel’s highest-density processor, will be available in a variety of form factors that cater to different customer needs.
The Xeon CPU Max Series offers 64 gigabytes of high-bandwidth memory (HBM2e) on the package, significantly increasing data transfer speed for HPC and AI workloads. Xeon CPU Max delivers up to 3.7x the performance of high-end 3rd Gen Intel® Xeon® Scalable processors in many real-world applications such as power and ground systems modelling.
In addition, the Data Center GPU Max Series packs over 100 billion transistors into a 47-tile package, providing new levels of throughput for demanding workloads such as physics, financial services, and life sciences. When paired with the Xeon CPU Max Series, the combined platform achieves 12.8 times higher performance than the previous generation when running the LAMMPS molecular dynamics simulator.
The Most Feature-Rich and Secure Xeon Platform Ever
Representing Intel’s greatest platform transformation to date, the 4th Gen Xeon is an acceleration marvel, combining up to four Intel 7-structured tiles in a single package, connecting using Intel EMIB (built-in multi-die interconnect bridge), and more advanced with DDR5. It is a manufacturing achievement that offers new features including higher memory bandwidth, higher I/O bandwidth with PCIe5.0 and Compute Express Link (CXL) 1.1 interconnect.
Security is at the heart of everything. With the 4th Gen Xeon, Intel offers the most comprehensive portfolio of confidential computing of any data center silicon provider in the industry, taking data security, regulatory compliance and data independence to the next level.
Intel; It remains the only silicon provider to offer application isolation for data center computing with Intel® Software Guard Extensions (Intel® SGX), which provides today’s smallest attack surface for stealth computing in private, public and cloud-to-end environments.
In addition, Intel’s new virtual machine (VM) isolation technology and Intel Trust Domain Extensions (Intel® TDX), ideal for moving existing applications to a confidential environment; It will be released with Microsoft Azure, Alibaba Cloud, Google Cloud and IBM Cloud.
Finally, 4th Gen Xeon’s modular architecture can be used for customer use cases or applications, from Intel’s mainstream general-purpose SKUs to purpose-built SKUs for cloud, database and analytics, networking, storage, and single-socket edge use cases. It allows it to offer a wide array of processors at around 50 SKUs targeted for
The 4th Gen Xeon processor family is On Demand and differs in number of cores, frequency, mix of accelerators, power and memory efficiency, suitable for target use cases and form factors that meet customers’ real-world needs.
4045
