PipeEdge: Pipeline Parallelism for Large-Scale Model Inference on Heterogeneous Edge Devices-Reference-Cited by-同舟云学术

PipeEdge: Pipeline Parallelism for Large-Scale Model Inference on Heterogeneous Edge Devices

Published:2022-08 Issue: Volume: Page:
ISSN:
Container-title:2022 25th Euromicro Conference on Digital System Design (DSD)
language:
Short-container-title:

Author:

Hu Yang¹,Imes Connor²,Zhao Xuanang³,Kundu Souvik¹,Beerel Peter A.¹,Crago Stephen P.²,Walters John Paul²

Affiliation:

1. Computer Engineering, University of Southern California,Ming Hsieh Department of Electrical,USA

2. Information Sciences Institute, University of Southern California,USA

3. Unaffiliated,China

Publisher

IEEE

Link

http://xplorestaging.ieee.org/ielx7/9996588/9996595/09996638.pdf?arnumber=9996638

Reference66 articles.

1. Fine-grained elastic partitioning for distributed dnn towards mobile web ar services in the 5g era;ren;IEEE Transactions on Services Computing,2021

2. DeepHome

3. Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding;han;ArXiv Preprint,2015

Cited by 3 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Distributed Edge Machine Learning Pipeline Scheduling with Reverse Auctions;2023 Eighth International Conference on Fog and Mobile Edge Computing (FMEC);2023-09-18

2. Quantpipe: Applying Adaptive Post-Training Quantization For Distributed Transformer Pipelines In Dynamic Edge Environments;ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP);2023-06-04

3. FactionFormer: Context-Driven Collaborative Vision Transformer Models for Edge Intelligence;2023 IEEE International Conference on Smart Computing (SMARTCOMP);2023-06