A Data and Model Parallelism based Distributed Deep Learning System in a Network of Edge Devices

计算机科学 GSM演进的增强数据速率 云计算 边缘计算 边缘设备 节点(物理) 启发式 仿真 分布式计算 加入 特征(语言学) 深度学习 并行计算 人工智能 哲学 程序设计语言 经济 工程类 操作系统 结构工程 经济增长 语言学
作者
Tanmoy Sen,Haiying Shen
标识
DOI:10.1109/icccn58024.2023.10230190
摘要

With the emergence of edge computing along with its local computation advantage over the cloud, methods for distributed deep learning (DL) training on edge nodes have been proposed. The increasing scale of DL models and large training dataset poses a challenge to run such jobs in one edge node due to resource constraints. However, the proposed methods either run the entire model in one edge node, collect all training data into one edge node, or still involve the remote cloud. To handle the challenge, we propose a fully distributed training system that realizes both Data and Model Parallelism over a network of edge devices (called DMP). It clusters the edge nodes to build a training structure by taking advantage of the feature that distributed edge nodes sense data for training. For each cluster, we propose a heuristic and a Reinforcement Learning (RL) based algorithm to handle the problem of how to partition a DL model and assign the partitions to edge nodes for model parallelism to minimize the overall training time. Taking advantage of the feature that geographically close edge nodes sense similar data, we further propose two schemes to avoid transferring duplicated data to the first-layer edge node as training data without compromising accuracy. Our container-based emulation and real edge node experiments show that our systems reduce up to 44% training time while maintaining the accuracy comparing with the state-of-the-art approaches. We also open sourced our source code.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
jeff完成签到,获得积分10
1秒前
merrylake完成签到 ,获得积分10
2秒前
Kong完成签到,获得积分10
3秒前
eric888应助小趴菜采纳,获得200
3秒前
4秒前
ZHANGXUEJUN完成签到,获得积分10
4秒前
蔡从安发布了新的文献求助10
4秒前
burn完成签到,获得积分10
4秒前
muzi完成签到,获得积分10
5秒前
chao完成签到,获得积分10
6秒前
day_on发布了新的文献求助10
7秒前
一一完成签到,获得积分10
8秒前
orixero应助Champ采纳,获得10
8秒前
一点完成签到 ,获得积分0
8秒前
ZYN完成签到 ,获得积分10
10秒前
搜集达人应助永远的阿科采纳,获得20
11秒前
SYLH应助蔡从安采纳,获得10
11秒前
13秒前
照镜子丫dorime完成签到,获得积分10
15秒前
奉雨眠完成签到,获得积分10
15秒前
埋头苦干科研完成签到,获得积分10
16秒前
闫栋完成签到 ,获得积分10
16秒前
稳重的秋天完成签到,获得积分20
18秒前
18秒前
坐雨赏花完成签到 ,获得积分10
18秒前
文献狗完成签到,获得积分10
18秒前
DMMM完成签到,获得积分10
19秒前
day_on完成签到,获得积分10
19秒前
车非笑完成签到,获得积分10
19秒前
20秒前
玛卡巴卡发布了新的文献求助10
20秒前
王二哈完成签到,获得积分10
21秒前
芝麻福福完成签到,获得积分10
21秒前
曼夭非夭完成签到,获得积分10
22秒前
wang完成签到,获得积分10
23秒前
xue完成签到 ,获得积分10
23秒前
zhang完成签到,获得积分10
24秒前
24秒前
小啊刘呀发布了新的文献求助10
25秒前
俍璟完成签到 ,获得积分10
25秒前
高分求助中
The Mother of All Tableaux Order, Equivalence, and Geometry in the Large-scale Structure of Optimality Theory 2400
Ophthalmic Equipment Market by Devices(surgical: vitreorentinal,IOLs,OVDs,contact lens,RGP lens,backflush,diagnostic&monitoring:OCT,actorefractor,keratometer,tonometer,ophthalmoscpe,OVD), End User,Buying Criteria-Global Forecast to2029 2000
Optimal Transport: A Comprehensive Introduction to Modeling, Analysis, Simulation, Applications 800
Official Methods of Analysis of AOAC INTERNATIONAL 600
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 588
Residual Stress Measurement by X-Ray Diffraction, 2003 Edition HS-784/2003 588
T/CIET 1202-2025 可吸收再生氧化纤维素止血材料 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3950021
求助须知:如何正确求助?哪些是违规求助? 3495348
关于积分的说明 11076451
捐赠科研通 3225877
什么是DOI,文献DOI怎么找? 1783346
邀请新用户注册赠送积分活动 867596
科研通“疑难数据库(出版商)”最低求助积分说明 800839