1. 南京邮电大学计算机学院,江苏,南京,210003
2. State Key Laboratory for Novel Software Technology at Nanjing University,Jiangsu,Nanjing,China,210093
3. 南京邮电大学计算机学院江苏南京,210003
4. State Key Laboratory for Novel Software Technology at Nanjing UniversityNanjingJiangsu 210093China
纸质出版:2009
移动端阅览
易 侃, 王汝传. 基于Nash均衡的网格多调度节点的任务调度算法[J]. 电子学报, 2009,37(2):329-333.
YI Kan, WANG Ru-chuan. Nash Equilibrium Based Task Scheduling Algorithm of Multi-schedulers in Grid Computing[J]. Acta Electronica Sinica, 2009, 37(2): 329-333.
目前网格任务调度算法主要是针对1×
n
型即单调度节点多资源的网格环境
而针对
m×n
型的网格环境研究较少.论文用
M/M/1排队系统对m×n
型网格环境建模
然后以每个调度节点调度任务的平均完成时间为优化目标
提出了
m×n
型网格环境任务调度的Nash均衡问题
并利用粒子群算法求得该Nash均衡解.通过仿真验证了该算法在单位时间内平均完成的任务数
网络平均负载
以及系统的平均负载上均优于基于均匀调度策略的调度算法.
At present
grid task scheduling Algorithms focus on 1×
n
type grid
namely one scheduler and
n
resources but neglect
m×n
type grid.We built a Grid model of
m×n
type grid using
M/M/1
queue system
and promoted the concept of task scheduling Nash equilibrium among multi-schedulers.The optimal objective of each scheduler is mean complete time per task.The Nash equilibrium took advantage of PSO to be solved.By simulations
we conclude that the new algorithm is better than the algorithm based on the mean scheduling strategies in mean finished task numbers per time
mean load of network and mean load of Grid resources.
0
浏览量
1571
下载量
5
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621