Using Dataflow to Optimize Energy Efficiency of Deep Neural Network Accelerators

The authors demonstrate the key role dataflows play in the optimization of energy efficiency for deep neural network (DNN) accelerators. By introducing a systematic approach to analyze the problem and a new dataflow, called Row-Stationary, which is up to 2.5 times more energy efficient than existing...

Full description

Saved in:
Bibliographic Details
Published in:IEEE MICRO 2017, Vol.37 (3), p.12-21
Main Authors: Yu-Hsin Chen, Emer, Joel, Sze, Vivienne
Format: Article
Language:English
Subjects:
Accelerators
Computer architecture
dataflow
Deep learning
deep neural network
Energy consumption
Energy efficiency
Energy management
Neural networks
Optimization
Power efficiency
Program processors
Radio frequency
Random access memory
spatial architecture
State of the art
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The authors demonstrate the key role dataflows play in the optimization of energy efficiency for deep neural network (DNN) accelerators. By introducing a systematic approach to analyze the problem and a new dataflow, called Row-Stationary, which is up to 2.5 times more energy efficient than existing dataflows in processing a state-of-the-art DNN, this work provides guidelines for future DNN accelerator designs.
ISSN:0272-1732
1937-4143
DOI:10.1109/MM.2017.54