基于商空间的层次式数据网格资源调度算法

发布时间：2018-12-27 06:37

【摘要】：为了解决传统数据网格调度算法在对层次式数据网格调度过程中出现的极易陷入局部最优值和收敛速度过慢的问题,将粒计算的思想引入到网格调度中,提出了一种基于商空间的层次式数据网格资源调度QSHDGRA(quotient space theory based hierarchical data grid resource allocation)算法。首先分析了层次式数据网格的特点,接着提出一种基于业务请求平均等待时间和网络与节点资源利用均衡度的调和函数的调度问题模型,随后设计了基于商空间的层次式最优资源调度算法。该算法的特点是可以在不同粒度上由粗至细地对网格业务进行调度,从而保证不同业务的QoS,并实现系统全局最优资源分配。仿真实验表明,算法可以显著地提升系统整体的吞吐率,具有更快的收敛速度,并具备线性扩展能力。
[Abstract]:In order to solve the problem that the traditional data grid scheduling algorithm is prone to fall into local optimal value and slow convergence speed in the process of hierarchical data grid scheduling, the idea of granular computing is introduced into grid scheduling. A hierarchical data grid resource scheduling QSHDGRA (quotient space theory based hierarchical data grid resource allocation) algorithm based on quotient space is proposed. Firstly, the characteristics of hierarchical data grid are analyzed, and then a scheduling problem model based on the average waiting time of service request and the balance between network and node resource utilization is proposed. Then a hierarchical optimal resource scheduling algorithm based on quotient space is designed. The characteristic of this algorithm is that the grid services can be scheduled from coarse to fine in different granularity, so that the QoS, of different services can be guaranteed and the global optimal resource allocation can be realized. Simulation results show that the algorithm can significantly improve the overall throughput of the system and has faster convergence speed and linear expansion ability.
【作者单位】：江苏大学计算机科学与通信工程学院;江苏大学信息化中心;
【基金】：十一五国家科技支撑计划基金资助项目(2010BAI88B00) 国家自然科学基金资助项目(61272074) 江苏省自然科学基金资助项目(BK2011464) 江苏省普通高校研究生科研创新计划基金资助项目(CXLX11_0591)~~
【分类号】：TP393.01

【相似文献】