CHIPSIM: A Co-Simulation Framework for Deep Learning on Chiplet-Based Systems

2025-10-29

AI Summary

To address the lack of high-fidelity, high-efficiency co-simulation methodologies for chiplet-based systems, this paper proposes the first deep learning task co-simulation framework enabling microsecond-level power and transient thermal awareness. The framework unifies modeling of computation, on-chip interconnect (NoI) communication, dynamic power dissipation, and 3D thermal diffusion—accurately capturing congestion, pipeline effects, and electro-thermal coupling. It supports parallel DNN execution simulation across both homogeneous and heterogeneous chiplet configurations, as well as diverse interconnect topologies. Experimental evaluation demonstrates a 340% improvement in accuracy over conventional simulation approaches, while maintaining flexibility and scalability. By enabling rapid, physics-informed assessment of system-level performance and thermal behavior, the framework provides an efficient and reliable tool for chiplet architecture design and optimization.

Technology Category

• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: (Computational) Cognitive Architectures
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Due to reduced manufacturing yields, traditional monolithic chips cannot keep up with the compute, memory, and communication demands of data-intensive applications, such as rapidly growing deep neural network (DNN) models. Chiplet-based architectures offer a cost-effective and scalable solution by integrating smaller chiplets via a network-on-interposer (NoI). Fast and accurate simulation approaches are critical to unlocking this potential, but existing methods lack the required accuracy, speed, and flexibility. To address this need, this work presents CHIPSIM, a comprehensive co-simulation framework designed for parallel DNN execution on chiplet-based systems. CHIPSIM concurrently models computation and communication, accurately capturing network contention and pipelining effects that conventional simulators overlook. Furthermore, it profiles the chiplet and NoI power consumptions at microsecond granularity for precise transient thermal analysis. Extensive evaluations with homogeneous/heterogeneous chiplets and different NoI architectures demonstrate the framework's versatility, up to 340% accuracy improvement, and power/thermal analysis capability.

Problem

• Simulating chiplet-based systems for deep learning with accuracy and speed
• Modeling network contention and pipelining effects in chiplet communications
• Profiling power consumption for precise thermal analysis in chiplets

Innovation

• Co-simulation framework for chiplet-based DNN systems
• Models computation and communication with network contention
• Profiles power consumption for precise thermal analysis