AMD made some noise Tuesday at Open Compute Summit V by unveiling its first ARM-based server chip using 28-nanometer processing technology.
During a keynote address, Andrew Feldman, corporate vice president and general manager of AMD's server business unit, said the AMD Opteron A1100 series, codenamed "Seattle," not only signals the chipmaker's commitment to becoming the "leader in ARM CPUs for servers" but the dawn of new era for data management.
"This is not a one-generation process," he said. "There will be generation after generation after generation and each will improve and be more closely tied to the software and hardware enclosure it lives in."
The AMD Opteron A1100 series processor supports four or eight CPU cores based on ARM's 64-bit Cortex-A57. The developer kit includes a Micro-ATX motherboard, the A1100 processor, four registered DIMM slots for up to 128GB of DDR3 DRAM, PCI Express connectors configurable as a single x8 or dual x4 ports and eight SATA ports.
The company also announced it will contribute to the Open Compute Project, an open-source hardware effort working to reduce datacenter costs and improve efficiencies, a new AMD Open CS 1.0 Common Slot design based on the AMD Opteron A-series processor that's compliant with the new Common Slot specification for partners integrators to build with immediately.
Company officials said the Opteron A-series development platform will be supported by a number of tools and applications including a standard UEFI boot and Linux environment based on the Fedora project.
"We are racing ahead in the ARM community," Feldman said. "We've passed youth and moved to the gangly adolescent phase for software" in the ARM ecosystem.
He added that ARM CPUs will play a "monstrous role" in tomorrow's datacenters, perhaps commanding a quarter of the server market by 2019.
"In the history of compute, smaller lower-priced CPUs have always won," he said.