• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Arabian Journal for Science and Engineering >Cloud Computing: A Multi-workflow Scheduling Algorithm with Dynamic Reusability
【24h】

Cloud Computing: A Multi-workflow Scheduling Algorithm with Dynamic Reusability

机译：云计算：具有动态可重用性的多工作流调度算法

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

Cloud computing provides a dynamic environment of well-organized deployment of hardware and software that are common in nature and the requirement for propping up heterogeneous workflow applications to realize high performance and improved throughput where the most demanding task is multiple workflow applications surrounded by their fixed deadline. These workflow applications consist of interconnected jobs and data. Nevertheless, hardly any initiations are tailored on multi-workflow scheduling exertion. These scheduling problems have been considered methodically in cloud atmosphere. Accessibility of the computing resources on the data center (DC) provides the exact time of execution of each process, whereas the execution time of every process within a workflow is pre-calculated in the majority of the existing multi-workflow scheduling problem. System overhead so far is an additional concern at the same time as dynamically generating virtual machines (VMs) with salvage them dipping the power eating. The aim of this paper is to reduce the execution time of every job and finalize the execution of all workflow within its deadline by producing VMs dynamically in DC and recycle them as necessary. We recommend a dynamic multi-workflow scheduling algorithm formally named as competent dynamic multi-workflow scheduling (CDMWS) algorithm. Simulation process describes one of the best algorithms so far in terms of performance among subsistent algorithm and moves toward a new era of multi-workflow relevance.
机译：云计算提供了一个动态的环境，该环境可以组织本质上很常见的硬件和软件，并且需要支持异构工作流应用程序以实现高性能和更高的吞吐量，其中最苛刻的任务是在固定期限内围绕多个工作流应用程序。这些工作流程应用程序由相互关联的作业和数据组成。但是，几乎没有针对多工作流计划应用量身定制的启动方式。这些调度问题已在云环境中被系统地考虑。数据中心（DC）上计算资源的可访问性提供了每个进程的确切执行时间，而在大多数现有的多工作流调度问题中，工作流中每个进程的执行时间都是预先计算的。到目前为止，与动态生成虚拟机（VM）并挽救它们消耗功率的同时，系统开销也是一个额外的问题。本文的目的是通过在DC中动态生成VM并根据需要对其进行回收，从而减少每个作业的执行时间，并在截止期限内最终完成所有工作流的执行。我们建议将动态多工作流调度算法正式命名为主管动态多工作流调度（CDMWS）算法。就持久性算法的性能而言，仿真过程描述了迄今为止最好的算法之一，并朝着多工作流相关性的新时代迈进。

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