After a series of delays extending back to late 2021, Intel finally formally announced its fourth-generation Xeon Scalable processors, which also had been known previously under the code-name Sapphire Rapids. The announcement also covered the Xeon CPU Max Series (code-named Sapphire Rapids HBM) and the Data Center GPU Max Series (codenamed Ponte Vecchio).
“The launch of 4th Gen Xeon Scalable processors and the Max Series product family is a pivotal moment in fueling Intel’s turnaround, reigniting our path to leadership in the data center and growing our footprint in new arenas,” said Sandra Rivera, Intel executive vice president and general manager of the Datacenter and AI Group.
This week’s announcement comes after the Xeon launched initially was delayed from late 2021 into 2022 and eventually to early 2023, although Intel did hold firm to a pledge made last November on Twitter to hold a Xeon launch event this week, so in that respect this week’s announcement was somewhat anti-climactic. Intel also had spent quite a bit of time in recent months detailing the expected performance benefits of the fourth-gen Xeon processors and how Intel achieved those improvements through accelerator techniques.
Intel said the new Xeon also will support new capabilities for AI, the cloud, the network and edge, including 5G deployments, and supercomputers. Regarding acceleration for AI and other applications, Intel claimed the new Xeon processors have the “most built-in accelerators of any CPU in the world to tackle customers’ most important computing challenges across AI, analytics, networking, security, storage and HPC.”
The Xeon announcement comes as industry observers such as Jack Gold, principal analyst at J. Gold Associates, have pointed out how competitors have challenged and made progress against Intel in the data center market as the delays with the fourth-generation Xeon have lingered on. The company also has lost a tremendous amount of value of the stock market, and even as it has laid out ambitious foundry plans has faced growing concerns about its ability to deliver new chips to market on time.
Intel may now be hoping that the Xeon's already-well publicized performance benefits can help it regain some traction and silence doubters. Compared to its own previous-generation Xeon processors, the fourth-generation delivers a 2.9x average performance per watt efficiency improvement for targeted workloads when utilizing built-in accelerators, and up to 70-watt power savings per CPU in optimized power mode with minimal performance loss. The benefits also come at a 52% to 66% lower total cost of ownership, the company said.
Intel said a new Optimized Power Mode can deliver up to 20% socket power savings with a less than 5% performance impact for selected workloads. That is an example of a broader effort by Intel to leverage innovations in air and liquid cooling reduce total data center energy consumption.
The fourth-generation Xeon processors also achieve up to 10x higher PyTorch real-time inference and training performance with built-in Intel Advanced Matrix Extension accelerators. Intel’s 4th Gen Xeon unlocks new levels of performance for inference and training across a wide breadth of AI workloads. The Xeon CPU Max Series expands on these capabilities for natural language processing, with customers seeing up to a 20x speed-up on large language models.