Thermal Implications of Non-Uniform Power in BSPDN-Enabled 2.5D/3D Chiplet-based Systems-in-Package using Nanosheet Technology

2025-08-04

AI Summary

Nanosheet-based 2.5D/3D chiplet system-in-package (SiP) architectures enable unprecedented power density, yet conventional thermal analysis—relying on uniform power distribution assumptions—significantly underestimates peak temperatures and fails to capture thermally distinct behaviors between front-side and back-side power delivery networks (FSPDN vs. BSPDN) under realistic non-uniform workloads. Method: We propose a high-resolution (5 μm) non-uniform power mapping thermal modeling framework to systematically quantify localized hotspot effects on SiP thermal distribution. Contribution/Results: Our analysis reveals that BSPDN exhibits pronounced thermal degradation under non-uniform loading—a critical deficiency entirely masked by uniform-power models. This work presents the first nanosheet-level experimental validation of the necessity for fine-grained, coupled power–thermal modeling in advanced SiPs. It establishes both theoretical foundations and methodological support for PDN architecture selection and thermal–electrical co-optimization in next-generation 3D-integrated systems.

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Machine Learning: Deep Neural Architectures and Foundation Models
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Security and Privacy: Large-scale security measurements

Abstract

Advances in nanosheet technologies have significantly increased power densities, exacerbating thermal management challenges in 2.5D/3D chiplet-based Systems-in-Package (SiP). While traditional thermal analyses often employ uniform power maps to simplify computational complexity, this practice neglects localized heating effects, leading to inaccuracies in thermal estimations, especially when comparing power delivery networks (PDN) in 3D integration. This work examines the thermal impact of non-uniform power distributions on SiPs utilizing frontside (FSPDN) and backside (BSPDN) power delivery approaches. Using high-resolution thermal simulations with non-uniform power maps at resolutions down to 5 micrometers, we demonstrate that uniform power assumptions substantially underestimate peak temperatures and fail to reveal critical thermal differences between BSPDN and FSPDN configurations in 3D scenarios. Our results highlight that BSPDN configurations in 3D, although beneficial in simplified uniform scenarios, exhibit pronounced thermal penalties under realistic, localized workloads due to limited lateral heat spreading. These findings emphasize the necessity of adopting fine-grained, workload-aware power maps in early-stage thermal modeling to enable accurate PDN assessment and informed thermal-aware design decisions in advanced nanosheet-based 3D SiP.

Problem

• Analyzing thermal effects of non-uniform power in 2.5D/3D SiPs
• Comparing BSPDN and FSPDN thermal performance in nanosheet systems
• Addressing inaccuracies from uniform power assumptions in thermal modeling

Innovation

• High-resolution thermal simulations with non-uniform power maps
• Comparison of BSPDN and FSPDN in 3D integration
• Workload-aware power maps for accurate thermal modeling