网格市场中基于成本计算的任务调度研究  

A cost-computing based task scheduling in grid market

作　　者：刘晓茜[1] 杨寿保[1] 郭磊涛[1] 申凯[1] 

机构地区：[1]中国科学技术大学计算机科学与技术系,合肥230026

出　　处：《中国科学院研究生院学报》2008年第3期379-385,共7页Journal of the Graduate School of the Chinese Academy of Sciences

基　　金：国家自然科学基金项目(60673172;60273041);国家高技术研究发展计划(863)项目(2006AA01A110)资助

摘　　要：现有网格环境中,在线调度策略主要着眼于资源的分配管理,往往以满足用户的各种资源请求为目的,而对于服务方关注不够.为增大服务方的收益,鼓励节点主动提供服务贡献资源,提出网格计算市场中基于成本计算的任务调度策略,根据用户提交任务的相关信息,计算接受任务的沉没成本和机会成本以决定是否接受任务,使得资源提供者和资源请求者都实现自身的经济目标,促使市场向健康稳定的方向发展.实验数据表明,该调度策略降低了服务方的成本,提高了服务方的收益,可以更有效的促使节点主动贡献自己的资源.In Grid environment, with a view to allocate and manage Grid resources, on-line schedulers aim at meeting all kinds of uses＇ resources requests, rather than paying enough attention to resources providers. In order to increase resources providers＇ profit and encourage sites to provide service actively, a cost-computing based task scheduling scheme is proposed, which calculates sunk cost and opportunity cost, based on relative information provided by users, to decide whether to accept the task, such that both resource providers and customers will achieve their economic goals, therefore, to promote the development of whole market more stably. The experiment shows that the scheduler reduces the cost of resources providers, increases their profit, and effectively encourages sites to offer more resources.

关 键 词：计算市场 在线调度 沉没成本 机会成本 收益 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]