Intel recently showcased a number of near-threshold voltage (NTV) techniques that promise to significantly increase processor energy efficiency by decreasing the supply voltage.
Essentially, the goal of NTV techniques is to enable extremely low supply voltages by using circuits that are extremely robust; tolerating variability and resilient against errors.
According to Silicon Valley tech guru David Kanter, the most impressive demonstration was run on a full x86 microprocessor (Claremont) that consumed as little as 2mW and was powered by a solar panel.
”Intel’s research on NTV is particularly interesting, because it promises a path to continue scaling voltage and decreasing dynamic power,” Kanter explained in a detailed analysis posted on RealWorldTech.
”NTV computing techniques are most applicable to highly parallel workloads. Generally, NTV is an ideal fit for HPC workloads and works very well for graphics, but not general purpose CPUs.”
Indeed, as Kanter notes, the world of HPC combines the favorable qualities of both the graphics and server markets.
”[Remember], power is the most significant constraint for HPC systems. Like GPUs, workloads are very parallel and performance is largely measured in terms of throughput. As such, reducing frequency to improve efficiency is an excellent choice,” he said.
”Accuracy is crucial for HPC, and a variable precision FPU fits well. However, a more advanced design would be needed, since HPC has significant double precision requirements.”
Kanter concluded that Intel’s work on near-threshold voltage was a “novel approach” which demonstrates substantially greater throughput and energy efficiency – at the cost of frequency and die area.
”The nature of NTV is uniquely suited to HPC workloads, which will significantly benefit from the energy advantages and can afford the associated costs.
“Therefore, it is likely that the first commercial implementations will appear in Intel’s HPC-oriented products, such as Knights Landing or successor projects. Other potential targets include Intel’s integrated GPUs for PCs, tablets and smartphones as well as SoC components such as DSPs and image processors,” he added.