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

TL;DR

CHIPSIM is a comprehensive co-simulation framework that accurately models computation, communication, and power consumption for deep learning on chiplet-based systems, enabling better design and analysis.

Contribution

It introduces a novel co-simulation framework that captures detailed system behaviors, including contention and thermal effects, with significant accuracy improvements over existing methods.

Findings

Achieves up to 340% accuracy improvement

Effectively models power and thermal profiles at microsecond granularity

Demonstrates versatility across various chiplet and NoI architectures

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.