AMD targets software scalability with embedded x86

AMD has stepped back into the market for embedded processors with a quad core design derived from its desktop accelerated processing units (APUs), which combine a general-purpose x86 subsystem with a graphics processor unit (GPU).

Although AMD still sells processors from the Geode line acquired from National Semiconductor, this is the company’s first foray of its own into embedded processors for more than a decade. The architectural choices that the company has made pitch the G-series embedded APU as a higher-horsepower alternative to Intel’s Atom family.

Arun Iyengar, vice president and general manager of AMD’s embedded business unit, said the company plans to become more “balanced” as it reduces its reliance on the desktop market. By the fourth quarter of 2014, the company aims to have 20 per cent of its business coming from embedded applications.

Fabbed on TSMC’s 28nm process, the processors are designed for a power envelope of 9W to 25W, depending on frequency and number of active processors. Some sub-9W designs are expected later with fewer cores. To save space and power, the cores share a 2Mbyte level-two cache instead of accessing their own independent stores.

Iyengar said the approach used by AMD offers greater scalability than existing x86 embedded processors. “They can use the same device across all of their products with a single software investment. That’s a big problem for these box makers: they have a high R&D spend on software. This value is resonating well with them.”

The current G series processors are based on AMD’s own take on the x86 architecture but, having signed an ARM licence to produce processors for the dense server market, a second range of Gs can be expected that use the 64bit ARM cores.