Cool-3D: An End-to-End Thermal-Aware Framework for Early-Phase Design Space Exploration of Microfluidic-Cooled 3DICs

📅 2025-03-10
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
Early-stage design space exploration (DSE) for 3D integrated circuits (3DICs) faces challenges including stringent thermal constraints, complex vertical stacking architectures, diverse cooling strategies, and insufficient support from pre-RTL tools. Method: This paper proposes the first end-to-end thermal-aware DSE framework, tightly integrating architectural simulation (gem5), power modeling (McPAT), and high-fidelity thermal simulation (HotSpot 7.0) at the pre-RTL stage. It natively supports microfluidic cooling modeling, customizable vertical stack configurations, and parametrization of process and cooling structures. An innovative non-parametric extensibility interface balances design-space coverage with modeling granularity. Contribution/Results: Evaluated on three case studies, the framework accurately captures inter-layer thermal coupling while significantly reducing iteration cost. Both the framework and experimental datasets will be open-sourced to advance co-optimization of 3DIC architecture and cooling.

Technology Category

Application Category

📝 Abstract
The rapid advancement of three-dimensional integrated circuits (3DICs) has heightened the need for early-phase design space exploration (DSE) to minimize design iterations and unexpected challenges. Emphasizing the pre-register-transfer level (Pre-RTL) design phase is crucial for reducing trial-and-error costs. However, 3DIC design introduces additional complexities due to thermal constraints and an expanded design space resulting from vertical stacking and various cooling strategies. Despite this need, existing Pre-RTL DSE tools for 3DICs remain scarce, with available solutions often lacking comprehensive design options and full customization support. To bridge this gap, we present Cool-3D, an end-to-end, thermal-aware framework for 3DIC design that integrates mainstream architectural-level simulators, including gem5, McPAT, and HotSpot 7.0, with advanced cooling models. Cool-3D enables broad and fine-grained design space exploration, built-in microfluidic cooling support for thermal analysis, and an extension interface for non-parameterizable customization, allowing designers to model and optimize 3DIC architectures with greater flexibility and accuracy. To validate the Cool-3D framework, we conduct three case studies demonstrating its ability to model various hardware design options and accurately capture thermal behaviors. Cool-3D serves as a foundational framework that not only facilitates comprehensive 3DIC design space exploration but also enables future innovations in 3DIC architecture, cooling strategies, and optimization techniques. The entire framework, along with the experimental data, is in the process of being released on GitHub.
Problem

Research questions and friction points this paper is trying to address.

Addresses early-phase design space exploration for 3DICs.
Integrates thermal-aware analysis with microfluidic cooling support.
Provides flexible and accurate 3DIC architecture modeling.
Innovation

Methods, ideas, or system contributions that make the work stand out.

Integrates gem5, McPAT, HotSpot for 3DIC design
Supports microfluidic cooling for thermal analysis
Enables non-parameterizable customization interface
🔎 Similar Papers
No similar papers found.
R
Runxi Wang
University of Michigan – Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai 200240, China
Z
Ziheng Wang
University of Michigan – Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai 200240, China
T
Ting Lin
University of Michigan – Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai 200240, China
J
Jacob M. Raby
University of Michigan – Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai 200240, China
Mircea R. Stan
Mircea R. Stan
Virginia Microelectronics Consortium (VMEC) Professor, University of Virginia
VLSIlow powerProcessing-in-Memorynanoelectronics
Xinfei Guo
Xinfei Guo
Shanghai Jiao Tong University
VLSIEDAReliabilityLow PowerMicroarchitecture