A site selection method of DNS using the particle swarm optimization algorithm. Issue 5 (26th December 2016)

• Record Type: Journal Article
• Authors: Liao, Yilan; Chen, Wenwen; Wu, Kaichao; Li, Dongyue; Liu, Xin; Geng, Guanggang; Su, Zheng; Zheng, Zheng
• Is Part Of: Transactions in GIS. Volume 21:Issue 5(2017)
• Page Start: 969
• Page End: 983
• Publication Date: 2016-12-26
• Subjects: DNS logs -- genetic -- multi‐objective particle swarm and particle swarm optimization algorithms -- site selection methods; Geographic information systems -- Periodicals; 910.285
• DOI: 10.1111/tgis.12244
• Languages: English
• ISSNs: 1361-1682

Abstract

Abstract: The Domain Name System (DNS) is an essential component of the functionality of the Internet. With the growing number of domain names and Internet users, the growing rate and number of visit quantity and analytic capacity of DNS are also proportional to the Internet users' size. This study (based on the analysis of access popularity and the distribution of massive DNS log data) aims to optimize the configuration of the DNS sites, which has become an important problem. The ArcGIS software is used to show the temporal and spatial distributions of visit source of DNS logs. This study also analyzes the influence of different sites and the dependence on DNS service in different regions of the world. This information is important to further decision‐making on new DNS site selection. This article proposes new DNS site selection solutions, using particle swarm and multi‐objective particle swarm optimization algorithms for one new site and multiple sites, respectively. The results from particle swarm optimization, genetic, and simulated annealing algorithms were compared and experimental results confirmed the correctness and effectiveness of the proposed methods. The proposed methods could also be extended to solve other layout related issues, such as onsite facility layout and road network optimization.