A 20-Year Retrospective on Power and Thermal Modeling and Management

TL;DR

This paper reviews 20 years of research on power and thermal modeling and management in processors, highlighting techniques, strategies, and future challenges to improve energy efficiency and reliability.

Contribution

It provides a comprehensive comparison of modeling techniques and management strategies, and discusses emerging challenges in power and thermal management.

Findings

Neural network-based power estimation techniques have advanced significantly.

Data-driven thermal modeling approaches are increasingly accurate.

Dynamic management strategies effectively balance performance and power consumption.

Abstract

As processor performance advances, increasing power densities and complex thermal behaviors threaten both energy efficiency and system reliability. This survey covers more than two decades of research on power and thermal modeling and management in modern processors. We start by comparing analytical, regression-based, and neural network-based techniques for power estimation, then review thermal modeling methods, including finite element, finite difference, and data-driven approaches. Next, we categorize dynamic runtime management strategies that balance performance, power consumption, and reliability. Finally, we conclude with a discussion of emerging challenges and promising research directions.