Paper Info
Title
Multi-node Scheduling Algorithm Based on Clustering Analysis and Data Partitioning in Emergency Management Cloud.
Abstract
Real-time processing is a key problem for big data analysis and processing, especially in emergency management. Strongly promoted by the leading industrial companies, cloud computing becomes increasingly popular tool for emergency management, that is emergency management cloud. How to make optimal deployment of emergency management cloud applications is a challenging research problem. The paper proposes a multi-node scheduling algorithm based on clustering analysis and data partitioning in emergency management cloud. First, the presented method divides the cloud nodes into clusters according to the communication cost between different nodes, and then selects a cluster for the big data analysis services. Second, the load balancing theory is used to dispatch big data analysis to these computing nodes in a way to enable synchronized completion at best-effort performance. At last, to improve the real-time of big data analysis, the paper presents a multi-node scheduling algorithm based on game theory to find optimal scheduling strategy for each scheduling node. Experimental results show the effectiveness of our scheduling algorithm for big data analytics in emergency management. © 2013 Springer-Verlag.
Year
DOI
Venue
2013
10.1007/978-3-642-39527-7_18
WAIM Workshops
Keywords
Field
DocType
big data,emergency management cloud,multi-node scheduling
Data mining,Software deployment,Load balancing (computing),Scheduling (computing),Computer science,Emergency management,Game theory,Cluster analysis,Big data,Distributed computing,Cloud computing
Conference
Volume
Issue
ISSN
7901 LNCS
null
16113349
Citations 
PageRank 
References 
2
0.45
9
Authors
3
Name
Order
Citations
PageRank
Qingchen Zhang1304.97
Zhikui Chen269266.76
Liang Zhao336648.82