传统关系型档案数据库系统的信息资源整合ETL过程中需要逐表存储.若资源目录系统有海量库表时,系统性能和效率将是无法承受的;若有新增资源目录事项加入时,需不断地在数据库中新建表单,相应底层适配读写查找软件都需要重新开发.由此造成数据管理和开发成本高居高不下,软件代码的可扩展性和兼容性不能满足实时应用的需求.为此,提出一种基于MapReduce思想的构建数据解析抽取函数表的搜索方法,以解决OLAP应用场景中关系数据的查询和系统维护问题.经实验证明具有一定应用价值.%The information resources for the traditional relational database file system integration requires each ETL storage and access table by table.If there is huge amounts of tables in the resource category system,it's unbearable for the system performance and efficiency,and if there is a new resource directory items to join,it's needs to create a new form in the database continually.So the corresponding software about lower-level adaptive software including reading, writing and searching,have to be re-developed.As a result, data management and development costs are so high that software code scalability and compatibility do not meet the needs of real-time applications.In this paper,we proposed a search method based on MapReduce to construct a data parsing and extraction function table to solve the query and system maintenance of relational data in OLAP application scenarios.The experiment proved that it had certain application value.
展开▼
