云计算环境下支持复杂并行业务的高铁数据中心关键技术研究

项目名称： 云计算环境下支持复杂并行业务的高铁数据中心关键技术研究

项目编号： No.61272029

项目类型： 面上项目

立项/批准年度： 2013

项目学科： 自动化技术、计算机技术

项目作者： 徐维祥

作者单位： 北京交通大学

项目金额： 61万元

中文摘要： 云计算在大规模异构资源优化管理方面的优势使其逐渐成为新一代主流计算模式。本课题以确保高速铁路安全，提升其预防维护能力、运营效率、服务质量为目标，面向云计算环境下高铁数据中心大规模复杂并行业务处理与海量分布数据管理的需求，分析实际业务与数据的特征及关联，探讨计算和存储紧密耦合的演化机理，基于本体、核心元数据及约束理论构建高铁数据模型。以此为基础，探索数据中心层次化混合存储架构，采用虚拟化及大规模数据集并行处理技术，结合模糊理论与动态多目标多约束决策理论，综合考虑负载均衡以及自适应机制，研究数据布局策略、调度算法以及中间数据容错机制。同时，针对跨地域密集布控的高铁传感监测设备回传的海量不确定流数据，寻求复杂高效查询处理算法。最终通过理论分析与仿真实验相结合的手段验证所设计的机制和算法,为下一步高铁数据中心的设计与管理提供新的理论和技术支撑。

中文关键词： 云计算；高速铁路；动态多目标规划；数据布局；数据中心

英文摘要： Cloud computing has gradually become a new generation of main stream computing model because of its advantages on large-scale heterogeneous resources optimization management. Aiming to assure safe operation of high-speed rail and improve prevention and maintenance ability, operation efficiency and service quality, this research projectis oriented to the demand of large-scale complex parallel business processing and massive distributed data management in high-speed rail data center under cloud computing environment. On this basis, we explore the hierarchical hybrid storage architecture of the data center, consider the load balancing and adaptive mechanisms comprehensively, and discuss the data layout strategy, scheduling algorithm, as well as intermediate fault tolerance mechanism, in adoption of virtualization and large-scale data sets of parallel processing technology combined with fuzzy theory and dynamic multi-objective multi-constraint decision theory. Meanwhile, design the complex and efficient continuous query processing algorithm for the mass uncertain flow data in the cross-regional intensive dispatched high-speed rail sensing monitoring equipment. Finally, It is verified that the designed algorithms and mechanisms with the combination of theoretical analysis and simulation results, provide new theoretic

英文关键词： cloud computing；high-speed rail；multiobjective；data placement；data center