融合遗传算法与蚁群算法动态网格任务调度算法研究.pdfVIP

融合遗传算法和蚁群算法动态网格任务调度算法研究1 孙玉涛 毕殿杰 （安徽财经大学信息工程学院计算机科学与技术系 安徽蚌埠 233041 ） （sunyt@ 联系电话： ） Integration of Genetic Algorithm and Ant Colony Algorithm for Dynamic Grid Task Scheduling Sun Yutao and Bi Dianjie (Information Engineering Department of Computer Science and Technology,Anhui University of Finance, Anhui Bengbu,233041) Abstract Grid computing is the rise of contemporary art in the fields of computer science to a very high value and value of academic research topic. The direction of future development of the Internet is the network number of idle computing resources, storage resources, and scientific instruments and other resources available to be exploited fully rational. How efficient use of grid resources, grid scheduling problem that will become a focus of the study, although in traditional distributed and parallel computing, there are many sophisticated task scheduling algorithm, but because of the new features of the grid, made it necessary to study new algorithms to solve new problems, such as the scheduling problem NP security, efficient scheduling algorithms, resource heterogeneity and resource allocation decisions, parallel and distributed and so on. Combination of grid task scheduling to the characteristics of genetic algorithms and ant colony optimization algorithm for parameter settings, the problem is transformed to the grid task scheduling problem up, proposed the integration of genetic algorithm and ant colony algorithm for grid task scheduling algorithm. Key words Grid task；Genetic Algorithm ；Ant colony algorithm ；task scheduling ；Dynamic integration 摘 要 网格计算是当今计算机科学领域最新兴起的一项有很高学术价值和应用价值的研 究课题。未来互联网的发展方向是将网络中众多闲置的计算资源、存储资源以及科学仪器 等可用资源充分合理的加以利用。如何高效地使用网格资源，即网格调度问题也随之成为 研究的重点，虽然在传统的分布式并行计算中有很多成熟的任务调度算法，但由于网格的 新特性，使得必须研究新的算法来解决一些新出现的问题，如调度问题的 NP 安全性，调