Content area

Abstract

Translate

For Open IoT, we have proposed Tacit Computing technology to discover the devices that have data users need on demand and use them dynamically and an automatic GPU offloading technology as an elementary technology of Tacit Computing. However, it can improve limited applications because it only optimizes parallelizable loop statements extraction. Thus, in this paper, to improve performances of more applications automatically, we propose an improved method with reduction of data transfer between CPU and GPU. We evaluate our proposed offloading method by applying it to Darknet and find that it can process it 3 times as quickly as only using CPU.

Details

1009240
Subject
Parallel processing;
Data transfer (computers);
Performance enhancement;
Computation;
Reduction;
Graphics processing units;
Electronic devices;
Central processing units--CPUs
URL
http://arxiv.org/abs/1811.03882
Title
Parallel processing area extraction and data transfer number reduction for automatic GPU offloading of IoT applications
Author
Yamato, Yoji; Noguchi, Hirofumi; Kataoka, Misao; Isoda, Takuma; Demizu, Tatsuya
Publication title
arXiv.org; Ithaca
Publication year
2018
Publication date
Nov 9, 2018
Section
Computer Science
Publisher
Cornell University Library, arXiv.org
Source
arXiv.org
Place of publication
Ithaca
Country of publication
United States
University/institution
Cornell University Library arXiv.org
Publication subject
Biology, Statistics, Physics, Mathematics, Engineering--Electrical Engineering, Computers--Computer Engineering, Business And Economics--Banking And Finance
e-ISSN
2331-8422
Source type
Working Paper
Language of publication
English
Document type
Working Paper
Publication history
 
 
Online publication date
2018-11-14
Milestone dates
2018-11-09 (Submission v1)
Publication history
 
 
   First posting date
14 Nov 2018
ProQuest document ID
2133085014
Document URL
https://www.proquest.com/working-papers/parallel-processing-area-extraction-data-transfer/docview/2133085014/se-2?accountid=208611
Full text outside of ProQuest
http://arxiv.org/abs/1811.03882
Copyright
© 2018. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Last updated
2019-08-26
Database
ProQuest One Academic