Product reviews, deals and the latest tech news

Intel introduces its 4th generation of Xeon Scalable Processors for cloud, data centre, and AI computing

The long-awaited fourth generation Xeon Scalable CPU from Intel has arrived (code-named Sapphire Rapids). Intel called the unveiling of the new Xeon at this week’s event one of the company’s “most important product releases to date.”

A new series of Intel Xeon central processing units (CPUs) called Sapphire Rapids HBM were unveiled. It’s a high-bandwidth memory x86-based processor designed to speed up HPC programmes without needing any changes on the part of the developers. Also new is Intel’s Data Center GPU Max Series, codenamed Ponte Vecchio, which was developed to enhance edge, cloud, network, and edge computing.

Many HPC applications may be accelerated without requiring any modifications to the source code thanks to Intel’s 4th Generation Xeon Max Series CPU, the first and only x86-based processor with high bandwidth memory.

The new CPUs are crucial for Intel and its partners not just because they boost performance in data centres and cloud infrastructures, but also because of other reasons. As a result of the holdups caused by production problems and then a fault identified late last year, AMD was able to increase its footprint in data centres.

Intel claims that its 4th Generation Xeon processors provide customers a wide range of control over power and performance, allowing them to optimise CPU use and so more effectively contribute to their sustainability objectives. Intel’s most eco-friendly CPUs for data centres are these.

Consumers of the 4th Generation Intel Xeon may expect an improvement in average performance per watt efficiency of up to 2.9x for specified workloads thanks to the processors’ built-in accelerators.

Intel claims that the 4th Generation Xeon series significantly broadens its purpose-built, workload-first approach. In addition, the company provides a series of processors designed for low-latency, high-performance network and edge applications.

There hasn’t been any other x86-based processor with high bandwidth memory until now, but the Xeon CPU Max Series has changed all that. Intel claims that the Data Center GPU Max Series, their highest-density CPU, would be available in a number of form factors to cater to a wide range of customer specifications.

The 64 GB of high-speed RAM included with the Xeon CPU Max Series greatly improves data throughput for HPC and AI programmes.

Amazon Web Services, Microsoft Azure, Google Cloud Platform, IBM Cloud, and Oracle Cloud are just few of the major cloud providers that have already implemented Intel’s latest 4th Gen Xeon Scalable CPU. Amazon VP David Brown revealed that the newest processors are live in EC2 on AWS.

Intel claims that the Xeon CPU Max Series is up to 3.7 times as fast as the best 3rd Generation Intel Xeon Scalable processors in some real-world uses, such as energy and earth system modelling.

Intel has commended the security improvements made to the 4th Generation Xeon Scalable CPU. This processor is compatible with Intel’s Software Guard Extensions (SGX) included in the company’s Xeon Scalable CPUs of the third generation. Intel SGX provides safe computing by lowering the attack surface in public, private, and cloud-to-edge environments.

The new 4th Generation Xeon has Intel’s Trust Domain Extensions (TDX), a virtual-machine isolation technology designed to facilitate the migration of applications to secure computing environments.