SIAM: Chiplet-based Scalable In-Memory Acceleration with Mesh for Deep Neural Networks

TL;DR

SIAM is a comprehensive benchmarking simulator for chiplet-based in-memory computing architectures, enabling scalable, flexible evaluation of deep neural network acceleration with significant energy-efficiency improvements over GPUs.

Contribution

This work introduces SIAM, a novel end-to-end simulation tool for evaluating chiplet-based IMC architectures, addressing scalability, flexibility, and performance analysis for deep learning acceleration.

Findings

SIAM accurately models chiplet-based IMC systems with integrated device, circuit, and network components.

Benchmarking shows significant energy-efficiency gains over GPUs for ResNet-50 on ImageNet.

SIAM enables efficient design space exploration for large-scale deep learning accelerators.

Abstract

In-memory computing (IMC) on a monolithic chip for deep learning faces dramatic challenges on area, yield, and on-chip interconnection cost due to the ever-increasing model sizes. 2.5D integration or chiplet-based architectures interconnect multiple small chips (i.e., chiplets) to form a large computing system, presenting a feasible solution beyond a monolithic IMC architecture to accelerate large deep learning models. This paper presents a new benchmarking simulator, SIAM, to evaluate the performance of chiplet-based IMC architectures and explore the potential of such a paradigm shift in IMC architecture design. SIAM integrates device, circuit, architecture, network-on-chip (NoC), network-on-package (NoP), and DRAM access models to realize an end-to-end system. SIAM is scalable in its support of a wide range of deep neural networks (DNNs), customizable to various network structures and configurations, and capable of efficient design space exploration. We demonstrate the flexibility, scalability, and simulation speed of SIAM by benchmarking different state-of-the-art DNNs with CIFAR-10, CIFAR-100, and ImageNet datasets. We further calibrate the simulation results with a published silicon result, SIMBA. The chiplet-based IMC architecture obtained through SIAM shows 130$\times$ and 72$\times$ improvement in energy-efficiency for ResNet-50 on the ImageNet dataset compared to Nvidia V100 and T4 GPUs.