微型电脑应用

针对当前海量的用户用电地址,以及地址管理的混乱问题,提出两种解决方案。一是借助Hadoop框架体系完成对海量数据的存储,同时借助MapReduce提高对数据的运算和处理能力;二是引入结构化管理模型,将用户用电地址分为10层,划分到社区和门牌号,同时引入不同地名的别称。最后构建用户用电地址知识库,并搭建Hadoop测试平台,对上述的试验方案进行验证。结果表明,通过本文的地址的结构化处理,可详细查看不同客户的地址,同时在数据处理方面也明显高于传统的系统。

Two solutions are proposed to solve the confusion of the current mass of user addresses and address management. Firstly, the Hadoop framework is used to store massive data, and MapReduce is used to improve the computing and processing ability of the data. Secondly, a structured management model is introduced to divide the user's electricity address into 10 layers, which are divided into the community and the door number. At the same time, different nicknames are introduced. Finally, the user address knowledge base is constructed, and the Hadoop test platform is built to verify the above test scheme. The results show that the addresses of different customers can be viewed in detail through the structured processing of addresses in this paper, and the data processing is obviously faster than the traditional system.