Thermal Modeling and Management of Liquid-Cooled 3D Stacked Architectures

Abstract

Abstract. 3D stacked architectures are getting increasingly attractive as they improve yield, reduce interconnect power and latency, and enable integrating layers manufactured with different technologies on the same chip. However, 3D integration results in higher temperatures following the increase in thermal resistances. This chapter discusses thermal modeling and management of 3D systems with a particular focus on liquid cooling, which has emerged as a promising solution for addressing the high temperatures in 3D systems. We first introduce a framework that is capable of detailed thermal modeling of the interlayer structure containing microchannels and through-silicon-vias (TSVs). For energy-efficient liquid cooling, we describe a controller to adjust the liquid flow rate to meet the current chip temperature. We also discuss job scheduling techniques for balancing the temperature across the 3D system to maximize the cooling efficiency and to improve reliability.

PODTherm-GP: A Physics-Based Data-Driven Approach for Effective Architecture-Level Thermal Simulation of Multi-Core CPUs

“…Thermal simulation is an important component for thermal [6], [13] and power [14], [15] managements which are two major strategies to minimize the coupling effects between chip temperature and power consumption and to prevent semiconductor chips from overheating. Hot spots in semiconductor chips can be predicted through accurate thermal simulation, and they can then be suppressed via the power management techniques, such as DVFS [6].…”

“…The reduced temperature in turn improves the power efficiency of semiconductor chip. In addition, the chip thermal profile predicted by thermal simulation is needed in thermal management to improve chip performance and reduce power consumption through exploring the cooling techniques [13], [15] or thermal-aware task scheduling [3]. The existing thermal modeling approaches for semiconductor chips are briefly discussed as follows.…”

Controlled Direct Liquid Cooling of Data Servers

“…[22] uses resistor-capacitor networks to capture heat conduction, generation and storage within a single chip. In addition, convective liquid cooling has been investigated for 3D stacked architectures where using air as the advection medium becomes inadequate due to the manifold increase in the power consumption [23], [24], [25]. Notice, however, that these approaches focus on a single package and disregard any dynamical interactions with other passive and active components.…”

The Thermal-Constrained Real-Time Systems Design on Multi-Core Platforms -- An Analytical Approach

“…The thermal problem has become the bottleneck in the design of future generations of high-performance computing systems [116,112,33,32,116,149,23,123,29,107,128,135].…”