摘要
针对云计算环境下高动态变化的数据库系统负载无法保障服务质量QoS(Quality of Service)的问题,提出一种自适应的数据库系统配置框架。该框架能够根据负载的动态变化,自动调整系统资源配置,从而保障用户要求的服务质量。框架的核心是基于分层排队网络模型LQNM(Layered Queue Network Model)的性能预测模型,用于指导搜索最优的资源配置,满足用户的性能需求。采用Hadoop搭建云计算环境,以TPC-H基准数据进行实验,实验结果显示,通过自适应配置框架对系统资源的调控,系统的QoS得到了很大提升。
In cloud computing environment, the database system workload with highly dynamic variations is unable to guarantee QoS （ Quality of Service）. In order to solve this problem, we proposed an adaptive configuration framework for database system. This framework can automatically adjust the allocation of system resources according to the dynamic change of workload, so that guarantees the QoS required by users. The core of the framework is an LQNM （layered queuing network model）-based performance prediction model used to guide the search for the optimal allocation of resources and to meet users＇ performance requirements. We used Hadoop to build a cloud computing envi- ronment, and carried out experiment with TPC-H benchmark data. Experimental results showed that system＇s QoS was greatly improved through the regulation on system resource by the adaptive configuration framework.
出处
《计算机应用与软件》
CSCD
2015年第9期49-53,共5页
Computer Applications and Software
基金
国家科技支撑项目子课题(2012BAH04F02-2)
山西省自然科学基金项目(2012011011-2)
山西省国际科技合作项目(2012081034-2)
关键词
云计算
数据库系统
分层排队网络模型
自适应配置
性能模型
Cloud computing Database system Layered queuing network model Adaptive configuration Performance model
