多维QoS指导的DBC最优算法

目前应用于网格的一些调度算法过于简单，不能满足对QoS的多样化需求，比如DBC（deadline and budget constrained）调度算法只支持两维的QoS需求，即截止期限（deadline）和预算（budge）。而对于真正的网格应用，用户与系统之间的交互应该加强，用户应能对提交的工作提出多种多样的QoS需求。在对传统DBC算法进行优化的基础上，提出了“多维QoS指导的DBC最优算法”，以确保搜索到所有满足用户需求的资源，不仅包括价格、时间最优，而且满足他们自己定义的专有QoS需求。为了测试这一最优算法，所做的模拟实验是以上海网格环境中各高性能结点的数据信息作为参数。

DBC optimization algorithm of guide to multi-dimensional QoS

Current algorithms like DBC(deadline and budget constrained) scheduling algorithms are too simple,for they only consider two dimensional QoS requirements(deadline and budget).Considering different users demanding services for different QoS requirements besides deadline and budget,a multi-dimensional QoS strategy is presented,which embeds QoS requirements information into the previous DBC optimization model in Resource Broker to satisfy user's multi-dimensional QoS requirements.The comparison between this algorithm and the conventional ones is demonstrated by simulating the resources of Shanghai grid testbed and scheduling tasks of dif-ferent deadline,budget,and QoS requirements scenarios.