Paper Info
Title
MapReduce-based optimization of overlay networks using particle swarm optimization
Abstract
An overlay network is a virtual network that is built on top of the real network such as the Internet. Cloud computing, peer-to-peer networks, and client-server applications are examples of overlay networks since their nodes run on top of the Internet. The major needs of overlay networks are content distribution and caching, file sharing, improved routing, multicast and streaming, ordered message delivery, and enhanced security and privacy. The focus of this paper is the optimization of overlay networks using a Particle Swarm Optimization (PSO) approach. However, since the ever growing need for more infrastructure causes the number of network nodes to grow significantly, the parallelization of the PSO approach becomes a necessity. In this paper, the MapReduce concept, proposed by Google, is adopted for the PSO approach in order to be able to optimize large-scale networks. MapReduce is easy to implement since it is based on the divide and conquer method, and implementation frameworks such has Hadoop allow for scalability and fault tolerance. Experiments of the MapReduce based PSO algorithm are performed to investigate the solution quality and scalability of the approach.
Year
DOI
Venue
2014
10.1145/2576768.2598269
GECCO
Keywords
Field
DocType
evolutionary computation,heuristic methods,overlay network optimization
Virtual network,Particle swarm optimization,Computer science,Computer network,Node (networking),Multicast,Overlay network,The Internet,Distributed computing,Cloud computing,Scalability
Conference
Citations 
PageRank 
References 
0
0.34
14
Authors
1
Name
Order
Citations
PageRank
Simone A Ludwig11309179.41