Touting a range of use cases from cloud computing to AI, Intel said the Xeon Scalable processors mark the "biggest data center advancement" the industry has seen in a decade, with up to 1.65 times higher performance power than its previous generation of chips.
The system aligns closely with Intel's data storage, networking and memory products. More broadly, the platform is part of Intel's pushback against rival chipmaker AMD -- which recently rolled out its latest Zen-based CPUs -- and an overall expansion of Intel's datacenter efforts.
Intel said more than 500,000 Xeon Scalable processors have already sold as part of its early ship program to major cloud players including Google and Amazon Web Services. Server vendors including Dell EMC and Hewlett Packard Enterprise are expected to unveil systems based on the new processors.
"Data center and network infrastructure is undergoing massive transformations to support emerging use cases like precision medicine, artificial intelligence and agile network services paving the path to 5G," said Navin Shenoy, EVP and GM of the Intel Data Center Group.
One of the biggest changes to the new Xeon Scalable chip family is the inclusion of mesh architecture, which allows data to flow directly between cores for improved performance. Intel said the chips support up to 28 cores and 56 threads per socket.
The chips also support up to 6 terabytes of system memory and can scale to support 2-socket through 8-socket systems and beyond, Intel said. Additional platform upgrades include support for AVX-512, as well as QuickAssist, Optane SSDs, and Intel Omni-Path Fabric.