基于语义搜索引擎的云资源调度  

Cloud resource scheduling based on semantic search engine

作　　者：黎明[1] 吴跃[1,2] 陈佳[1,3] 

机构地区：[1]电子科技大学计算机科学与工程学院,成都611731 [2]电子科技大学无锡研究院,江苏无锡214135 [3]美国加州大学欧文分校电子工程学院,加州92697

出　　处：《计算机应用研究》2015年第12期3735-3737,3749,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(61103111);国家科技型中小企业创新基金资助项目(14C26213201050);无锡市政产学研合作项目(CMES3C1101)

摘　　要：针对现有云资源调度存在的非自动调度、云资源的非优化使用等问题,提出了采用语义搜索引擎(semantic search engine,SSE)实现云资源调度的有效方法。SSE是由美国加州大学欧文分校与电子科技大学联合开发的,它利用云资源与云用户请求的语义信息构造相应本体,进行资源的较优匹配,自动、快速地为云用户需求建立一个优化的解决方案。首先,给出了基于SSE的云资源调度体系结构;其次,给出了SSE的体系结构、SQDL&SCDL匹配器的本体构造方法。仿真实验表明,基于SSE的云资源调度比传统的云资源调度的点对点通信性能提高了约15%,子任务执行时间减少了约18%,表明它是一种可行且高效的云资源调度方法。None of existing cloud systems is both being an automated scheduling and considering the optimal usage of resources. To address these problems, this paper proposed a cloud resource scheduling strategy using semantic search engine （SSECRS）. SSE was a new type of service search engine developed by University of California, Irvine and University of Electronic Science and Technology of China. SSE took full advantage of the semantic information between cloud resources and cloud users＇ requirements to construct the appropriate ontology to complete the appropriate match. SSE built an optimal solution according to users＇ requirements automatically and quickly. Firstly this paper presented the architecture of cloud resource scheduling based on SSE as well as the architecture of SSE. Then it described the ontology construct in SQDL ＆ SCDL match. The experiment shows point to point communication performance in SSECRS improves about 15% and reduces about 18% subtasks execution time comparing to traditional cloud resource scheduling（TCRS）. Hence it is a feasible and efficient cloud resource scheduling method.

关 键 词：云计算 资源调度 语义搜索引擎 本体 对象关系数据库 

分 类 号：TP393.07[自动化与计算机技术—计算机应用技术]