Modeling and Analysis of Hadoop MapReduce Systems for Big Data Using Petri Nets

Information technological advances have significantly increased large volumes of corporate datasets, which have also created a wide range of business opportunities related to big data and cloud computing. Hadoop is a popular programming framework used for the setup of a cloud computing system. The M...

Full description

Saved in:
Bibliographic Details
Published inApplied artificial intelligence Vol. 35; no. 1; pp. 80 - 104
Main Authors Chiang, Dai-Lun, Wang, Sheng-Kuan, Wang, Yu-Ying, Lin, Yi-Nan, Hsieh, Tsang-Yen, Yang, Cheng-Ying, Shen, Victor R. L., Ho, Hung-Wei
Format Journal Article
LanguageEnglish
Published Philadelphia Taylor & Francis 02.01.2021
Taylor & Francis Ltd
Taylor & Francis Group
Subjects
Big Data
Cloud computing
Data analysis
Petri nets
Online AccessGet full text

Cover

More Information
Summary:Information technological advances have significantly increased large volumes of corporate datasets, which have also created a wide range of business opportunities related to big data and cloud computing. Hadoop is a popular programming framework used for the setup of a cloud computing system. The MapReduce framework forms a core of the Hadoop program for parallel computing and its parallel framework can greatly increase the efficiency of big data analysis. This paper aims to adopt a Petri net (PN) to create a visual model of the MapReduce framework and to analyze its reachability property. We present a real big data analysis system to demonstrate the feasibility of the PN model, to describe the internal procedure of the MapReduce framework in detail, to list common errors and to propose an error prevention mechanism using the PN models in order to increase its efficiency in the system development.
ISSN:0883-9514
1087-6545
DOI:10.1080/08839514.2020.1842111