一种求解面向服务软件部署优化问题的多目标蚁群算法

来源期刊:中南大学学报(自然科学版)2017年第9期

论文作者:应时 李琳 董波

文章页码:2376 - 2388

关键词:面向服务软件;部署优化;多目标蚁群算法;性能

Key words:service-oriented software; deployment optimization; multi-objective ant colony algorithm; performance

摘    要:基于根据动态变化的外部环境调整面向服务软件的部署方案是提升其运行性能、降低运行成本的一种有效途径,提出一种基于多目标蚁群算法的MACO-DO,以便在自动为面向服务软件寻找一组在性能和成本之间作出最优权衡的部署方案。MACO-DO算法是对传统多目标蚁群算法的一种改进,引入摒弃精英解策略以避免算法早熟收敛,设计1个局部搜索过程以加快获得可行解的过程。在Case 1,Case 2和Case 3共3种不同规模的模拟案例上将提出的MACO-DO算法与P-ACO算法和NSGA-Ⅱ算法进行对比。研究结果表明:MACO-DO算法在求解问题上具有更好的性能。

Abstract: Considering that optimizing the deployment architecture of service-oriented software according to the changes in runtime environment is an efficient way to improve its performance and reduce its cost, a multi-objective ant colony algorithm MACO-DO was proposed for the automatic exploration of search space to find a set of pareto optimal deployment architectures for a service-oriented application quickly. MACO-DO is an improved version of the traditional multi-objective ant colony algorithm. A strategy of discarding elitist solutions was introduced to prevent the algorithm from prematuring convergence and a local search procedure was designed to accelerate the process of obtaining feasible solutions. A series of experiments were implemented on three simulated cases (i.e. Case 1, Case 2 and Case 3) of different sizes to verify the efficiency of MACO-DO and it was compared with P-ACO and NSGA-Ⅱ. The results show that MACO-DO has better performance than others on the considered problem.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号