基于QoS的网格服务匹配随机Petri网模型

提出了随机Petri网网格服务质量模型,对网格调度实体和过程进行抽象和简化;图形化地反映当前网格资源分派状况,利用经验学习的方法对网格服务进行匹配,并对网格服务质量进行了定量分析.为网格服务质量评价提供了一个定性和定量的评价方法,对于减少网格服务资源冲突、提高服务质量具有重要意义.

A Stochastic Petri Net Model for QoS Based Grid Service Matching

It is very important to match well resources with requests to improve the utilization rate of grid resources and QoS for users.A stochastic Petri net model is thus proposed for QoS based grid service matching to abstract and simplify the grid scheduling entities/processes.The model takes advantages of graphical and mathematical methods to simulate the grid resource allocation processes and match the resource supply with service demand more precisely and flexibly by the method of learning by experiences.Furthermore,a quantitative analysis of the performance of the model is given.The model provides both a qualitative and a quantitative evaluation method for grid QoS.This is important to the reduction of the conflicts between the resources and the services and the improvement of grid QoS.