To get a bigger bang for the IT buck, IT managers like to optimize infrastructure for targeted applications and workloads. As I noted in a recent Intel Chip Chat, one way to do this is by creating CPUs targeted at certain types of workloads. That was the case with Centeron, the code name for the Intel® Atom™ S1200 processor family, launched in 2012.
Centerton was the industry’s first 64-bit sub-6-watt system on a chip (SoC). This low-power processor expanded our server product line, building on the Intel® Xeon® processor family to support a new class of high-density microservers. Now we’re coming up on the launch of Avoton, the follow-on to Centeron, which will debut later this year.
Great stuff, all the way around. But at Intel, we know that processors are only part of the optimization story. We like to look at the entire data center infrastructure to make sure we’re optimizing across the board.
With those thoughts in mind, we are collaborating with leading cloud service providers to rethink rack designs to make them more modular and efficient. Specifically, we’re developing a reference design that uses a suite of innovative technologies in a disaggregated rack-scale architecture.
This mix includes Intel Xeon processors and Intel Atom SoCs for servers, storage and networking; Intel® Ethernet switch silicon for distributed input/output; and Intel® Silicon Photonics Technology for high-speed interconnects.
For a closer look at this effort, stop into the Intel IT Center.
Raejeanne Skillern is Intel’s director of marketing for cloud computing.