Temperature Regulation in Multicore Processors Using Adjustable-Gain Integral Controllers

TL;DR

This paper introduces an adjustable-gain integral controller for temperature regulation in multicore processors, enabling fast, accurate tracking despite model uncertainties and time-varying conditions.

Contribution

It presents a novel feedback law with an adjustable gain designed for nonlinear, time-varying models, improving temperature control efficiency and robustness.

Findings

Fast and accurate temperature tracking demonstrated in simulations

Outperforms fixed-gain controllers in response speed and accuracy

Effective in handling model uncertainties and dynamic conditions

Abstract

This paper considers the problem of temperature regulation in multicore processors by dynamic voltage-frequency scaling. We propose a feedback law that is based on an integral controller with adjustable gain, designed for fast tracking convergence in the face of model uncertainties, time-varying plants, and tight computing-timing constraints. Moreover, unlike prior works we consider a nonlinear, time-varying plant model that trades off precision for simple and efficient on-line computations. Cycle-level, full system simulator implementation and evaluation illustrates fast and accurate tracking of given temperature reference values, and compares favorably with fixed-gain controllers.