SoftMemoryBox II: A Scalable, Shared Memory Buffer Framework for Accelerating Distributed Training of Large-Scale Deep Neural Networks

SoftMemoryBox II: A Scalable, Shared Memory Buffer Framework for Accelerating Distributed Training of Large-Scale Deep Neural Networks

Distributed processing using high-performance computing resources is essential for developers to train large-scale deep neural networks (DNNs). The major impediment to distributed DNN training is the communication bottleneck during the parameter exchange among the distributed DNN training workers. T...

Full description

Bibliographic Details
Main Authors: Shinyoung Ahn, Eunji Lim
Format: Article
Language:English
Published: IEEE 2020-01-01
Series:IEEE Access
Subjects:
High-performance computing
distributed computing
SoftMemoryBox II
shared memory buffer
deep neural networks
distributed deep learning
Online Access:https://ieeexplore.ieee.org/document/9260142/

Internet

https://ieeexplore.ieee.org/document/9260142/

Similar Items