网格资源协同分配死锁处理问题研究  

Study on Deadlock Resource Co-allocation in Grid

作　　者：张横云[1] 段金蓉[2] 

机构地区：[1]四川科技职工大学,四川成都610101 [2]绵阳师范学院,四川绵阳621000

出　　处：《现代电子技术》2008年第16期112-115,共4页Modern Electronics Technique

摘　　要：网格中资源协同分配是资源组织和调度的一个重要组成部分,如何检测应用之间的死锁是资源协同分配过程中需要解决的重要问题。通过对网格中死锁原因的分析,对死锁的特点进行描述。提出基于Agent的网格资源协同分配死锁处理算法,并对算法进行实验验证。实验证明使用该方法不仅能够检测解除应用资源分配过程中的死锁,与其他方法相比,还能获得好的资源分配性能。Applications designed to execute on grid frequently requires the co-allocation. In particular,the problem of co-allocating grid resources that span multiple administrative domains are complicated by the possibility of deadlock. Motivated by this, the reason for application deadlock is analyzed and the deadlock in grid is characterized. Further, a new algorithm for co-allocation of grid resources is developed based on Agent dealing with deadlocks. At last, algorithm is studied by experiment. Experimental results demonstrate that the proposed method yields a significant performance improvement over the other existing deadlock prevention method.

关 键 词：网格技术 资源协同分配 死锁 分配性能 

分 类 号：TP393.7[自动化与计算机技术—计算机应用技术]