...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of information science and engineering >A Cost-Optimized GA-Based Heuristic for Scheduling Time-Constrained Workflow Applications in Infrastructure Clouds Using an Innovative Feasibility-Assured Decoding Mechanism
【24h】

A Cost-Optimized GA-Based Heuristic for Scheduling Time-Constrained Workflow Applications in Infrastructure Clouds Using an Innovative Feasibility-Assured Decoding Mechanism

机译：一种基于成本优化的基于启发式的启发式算法，使用创新的可行性保证解码机制在基础架构云中调度时间受限的工作流应用程序

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Recently, cloud computing has emerged as a realization of utility computing in which users have to pay for the utilized resources. Besides, there is a growing request for automation of current scientific and business applications in the form of workflows in cloud environments. While users wish to pay as less monetary cost as possible for executing their workflows, the time-constrained nature of this type of applications is a barrier to cost minimization. Solving this problem requires developing new efficient and customized scheduling algorithms. To this end, in this paper, first we have formulated a cloud-customized task-resource mapping which is exploited as the cost function of our Genetic Algorithm (GA)-based scheduling method. Also, by using indirect chromosome representation scheme and proposing a novel genotype-to-phenotype mapping (GPM), the algorithm guarantees the feasibility of the solutions and removes the restrictions imposed by evolutional operators and overheads of repair phases. Moreover, a key property of our method, called neutrality, strongly improves the quality of the solutions and the convergence rate. The results of experiments done on some real-world workflow benchmarks show that monetary cost of the solutions found by our algorithm outperform those of some recently successful scheduling algorithms. Moreover, the run time needed for the proposed method to produce solutions is in the order of seconds which demonstrates its quickness compared to other mentioned algorithms.
机译：近来，云计算已经出现为效用计算的实现，其中用户必须为所利用的资源付费。此外，对以云环境中的工作流形式实现的当前科学和业务应用程序自动化的需求不断增长。尽管用户希望为执行其工作流付出尽可能少的金钱成本，但此类应用程序受时间限制的性质是成本最小化的障碍。解决此问题需要开发新的高效和定制的调度算法。为此，在本文中，我们首先制定了云定制的任务-资源映射，该映射被用作基于遗传算法（GA）的调度方法的成本函数。此外，通过使用间接染色体表示方案并提出一种新的基因型到表型作图法（GPM），该算法保证了解决方案的可行性，并消除了进化算子和修复阶段开销所带来的限制。此外，我们方法的一个关键特性（称为中立性）极大地提高了解决方案的质量和收敛速度。在一些实际的工作流基准上进行的实验结果表明，我们的算法发现的解决方案的货币成本要优于某些最近成功的调度算法。而且，所提出的方法产生解决方案所需的运行时间为几秒左右，这证明了与其他提到的算法相比的快速性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号