文章基本信息

标题：HYBRID GENETIC VARIABLE NEIGHBORHOOD SEARCH BASED JOB SCHEDULING WITH DUPLICATION FOR CLOUD DATA CENTERS
本地全文：下载
作者：RACHHPAL SINGH ; KARANJIT SINGH KAHLON ; GURVINDER SINGH 等
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：22
页码：6204
出版社：Journal of Theoretical and Applied
摘要：Background/Objectives: Scheduling is one of the important way to provide high availability of processors to cloud users. Majority of scheduling approaches are NP-Hard. Therefore, meta-heuristics techniques are required to schedule the jobs on virtual machines (VMs). Meta-heuristic techniques usually suffer from inter-processor communication issues as well as premature convergence and global optima. Methods: To handle these issues, hybrid scheduling technique was proposed using Genetic Algorithm (GA) and Variable Neighborhood Search (VNS) with Task Duplication (TD). Thus, proposed technique can reduce the inter-processor scheduling overheads among high-end servers. Results: To attain the objectives of the proposed approach, the cloud based model was designed by considering well-known Fast Fourier Transformation (FFT) problem using Directed Acyclic Graph (DAG). A simulation environment was designed to implement the proposed technique. Extensive experiments have shown that the proposed technique outperforms over available techniques regarding Makespan, Speedup, and Efficiency. Conclusion: From a comparative analysis of existing and scheduling techniques it has been found that the mean reduction in makespan is 7.07%. The comparative studies have demonstrated that the mean improvement of proposed technique over other techniques concerning efficiency is 0.031%.
关键词：Cloud Environment; Task Duplication; Variable Neighborhood Search; Genetic Algorithm; Directed Acyclic Graph