An Evaluation of Edge TPU Accelerators for Convolutional Neural Networks
Edge TPUs are a domain of accelerators for low-power, edge devices and are widely used in various Google products such as Coral and Pixel devices. In this paper, we first discuss the major microarchitectural details of Edge TPUs. Then, we extensively evaluate three classes of Edge TPUs, covering different computing ecosystems, that are either currently deployed in Google products or are the product pipeline, across 423K unique convolutional neural networks. Building upon this extensive study, we discuss critical and interpretable microarchitectural insights about the studied classes of Edge TPUs. Mainly, we discuss how Edge TPU accelerators perform across convolutional neural networks with different structures. Finally, we present our ongoing efforts in developing high-accuracy learned machine learning models to estimate the major performance metrics of accelerators such as latency and energy consumption. These learned models enable significantly faster (in the order of milliseconds) evaluations of accelerators as an alternative to time-consuming cycle-accurate simulators and establish an exciting opportunity for rapid hard-ware/software co-design.
Code (1)
Methods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Exploration of Unary Arithmetic-Based Matrix Multiply Units for Low Precision DL Accelerators
General matrix multiplication (GEMM) is a fundamental operation in deep learning (DL). With DL moving increasingly toward low precision, recent works have proposed novel unary GEMM designs as an alternative to convention…
Revealing CNN Architectures via Side-Channel Analysis in Dataflow-based Inference Accelerators
Convolutional Neural Networks (CNNs) are widely used in various domains, including image recognition, medical diagnosis and autonomous driving. Recent advances in dataflow-based CNN accelerators have enabled CNN inferenc…
Autonomous DrivingMedical DiagnosisSide Channel AnalysisLW-GCN: A Lightweight FPGA-based Graph Convolutional Network Accelerator
Graph convolutional networks (GCNs) have been introduced to effectively process non-euclidean graph data. However, GCNs incur large amounts of irregularity in computation and memory access, which prevents efficient use o…
CPUGPUQuantizationPhotonic Reconfigurable Accelerators for Efficient Inference of CNNs with Mixed-Sized Tensors
Photonic Microring Resonator (MRR) based hardware accelerators have been shown to provide disruptive speedup and energy-efficiency improvements for processing deep Convolutional Neural Networks (CNNs). However, previous …
Bifrost: End-to-End Evaluation and Optimization of Reconfigurable DNN Accelerators
Reconfigurable accelerators for deep neural networks (DNNs) promise to improve performance such as inference latency. STONNE is the first cycle-accurate simulator for reconfigurable DNN inference accelerators which allow…