欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
动态网格环境中用户截止时间保障是工作流调度问题的一个挑战.利用随机服务模型来描述网格资源的动态处理能力及其动态负载压力,提出了截止时间满意度的概念和工作流截止时间满意度的计算方法.将以DAG图形式表示的任务执行关系转换为以数值表示的任务执行优先级,并根据最大截止时间满意度优先的思想,确定执行工作流子任务的候选资源;将工作流全局截止时间划分问题描述为一个约束下的非线性规划问题并通过已有方法求解该问题,提出了一种截止时间满意度增强的工作流调度算法（DSESAW）.仿真实验采用实际网格应用和系统数据来验证所提出算法的性能表现,实验结果表明新算法在网格环境的自适应性和用户截止时间保障方面优于其他两种实际网格系统中的调度算法.
英文摘要：
In grid,users usually pay more attention to the execution time of workflow than other QoS metrics.Consequently how to effectively guarantee meeting users＇ deadline requirements is a challenging problem for workflow scheduling in dynamic grid environment.Stochastic service model is utilized to describe dynamic processing capacity of grid resource and its dynamic workloads.The concept of deadline satisfaction degree（DSD） is defined and a corresponding calculation method for deadline satisfaction degree of workflow（DSDW） is provided.The task precedence relations represented in a DAG are converted into task execution priorities represented in numbers based on task length,and then the candidate resource for each task in the workflow is selected based on the rule of maximizing DSD.The deadline distribution is modeled as a non-linear programming problem with constraints and resolved with an interior point algorithm.A deadline satisfaction enhanced scheduling algorithm for workflow（DSESAW）,which includes resource selection and overall deadline distribution,is put forward finally.The extensive simulations using real-world workflow application and grid system are made to validate this algorithm.The experimental results show that this scheduling algorithm achieves better performance than other two algorithms used in real grid system on adaptation to dynamic grid environment and user＇s deadline guarantee.
同期刊论文项目
同项目期刊论文
期刊信息
位置：