Intel Unveils New Xeon Chips to Power AI Workloads

Intel has unveiled its next-gen Xeon scalable processors built to power AI and edge applications and supercomputers.

The chip-making giant showcases three new Xeon lines – the 4th Gen Intel Xeon, code-named Sapphire Rapids, the Intel Xeon CPU Max Series, code-named Sapphire Rapids HBM, and the Intel data center GPU Max Series, code-named Ponte Vecchio.

According to Intel, the 4th Gen Xeon processors are powerful but also the company’s “most sustainable data center processors.”

The tech giant also claims the chips improve AI inferencing and training by 10 times and three times for data analytics workloads.

The new processors have an entirely new microarchitecture – with increased core counts and built-in accelerators designed to achieve better performance. Intel said that the line achieves five times better performance for memory bandwidth against previous generations.

A long list of big names including Ericsson, Fujitsu, Google Cloud and HPE, are listed by Intel as customers for the new chips.

The new chips will drive “faster time to value and powering their pace of innovation,” according to Sandra Rivera, Intel executive vice president and general manager of the company’s data center and AI group.

The 4^th-Gen Xeons replaces the Ice Lake product line. Released two years ago, Ice Lake suffered several delays due to manufacturing issues concerning the chip’s 10nm nodes – forcing the company to use older 14nm nodes instead. The 10nm node was eventually rectified and placed into the Intel 7 line, released last year.

The new line of Xeons is optimized for AI and high-performance computing, SaaS, and 5G and networking. And according to Intel, 4^th Gen customers can get almost three times the average performance per watt efficiency improvement for targeted workloads when utilizing built-in accelerators.