摘要
随着OGSA(open grid service architecture)的提出,在网格环境中,如何进行服务的选择是一个热点问题.对于DAG(directed acyclic gragh)描述的网格工作流模型,考虑时间和费用两个Qos(quality of service)参数,即满足用户截止期的前提下,为工作流选择尽可能便宜的服务.对于一般遗传算法的求解问题时出现的早熟和退化现象,采用了免疫算子和遗传算子相结合的免疫遗传算法来进行搜索全局解,并且加入了自适应遗传交叉算子和疫苗提取来提高搜索能力.仿真试验证明文章的算法具有较好的解空间搜索性能.
With the OGSA proposed, how to select a grid service is growing a hot issue. Considering the Qos parameters of time and cost in grid workflow model depicted by DAG, that means to meet the user＇s deadline, the cheapest possible service is selected for workflow. As the appearance of prematurity and degradation in genetic algorithm, the genetic operator and immune operator are combined to used in this paper in order to search for global solutions. Moreover, self-adapted genetic cross operator and vaccine extraction are adopted to improve the search effeciency. The simulation experiement shows the algorithm proposed in this paper has better search performance in the solution space.
出处
《浙江工业大学学报》
CAS
北大核心
2010年第6期673-678,共6页
Journal of Zhejiang University of Technology
