【作者】 张哲；

【导师】 邓米克；

【作者基本信息】 北京工业大学 ， 计算机科学与技术， 2015， 硕士

【摘要】 近年来,我国通用航空产业在规模和作业量上取得了巨大进步,随着2015年低空空域的进一步开放,通用航空产业将进入发展的快车道。虽然我国通用航空产业发展潜力巨大,但是信息化水平远远落后于发达国家。大多数通用航空公司的数据管理还停留在纸质记录的原始阶段,部分通用航空公司虽然有自己的信息管理系统,但由于数据分散存储于各个子业务系统中,加上各通用航空公司之间地理上的隔离,数据无法实现有效共享,不仅存储效率低,而且数据的可靠性与安全性都得不到保障,更无法有效挖掘数据中蕴含的有用信息。因此,通用航空信息化水平低,信息管理水平严重滞后已成为我国通用航空产业发展的瓶颈。本文针对我国通用航空产业信息化面临的问题与实际需求,结合MongoDB以及REST自身的技术特点,使用Hadoop数据分析平台,构建通航云数据中心。云数据中心将会对通航公司产生的数据进行统一集中管理,为通航公司之间以及通航公司与用户之间搭建高效可靠的信息通道,提高飞行效率,保障飞行安全,有效提高通用航空信息化水平。本文主要工作与创新点如下:1)分析云数据中心相关技术。深入研究云计算架构与核心服务,MongoDB和REST的基本概念与特点以及MapReduce编程模型的实现原理。2)系统研究了通航云数据中心的需求,明确了系统的功能。深入分析了云数据中心的核心业务,确定了各个业务的数据处理流程。3)设计与实现了基于MongoDB与REST的通航云数据中心系统。系统使用NoSQL类型的文档数据库MongoDB存储数据,使用REST API为通航公司的PC端和移动端业务系统提供统一的数据访问服务。系统将MongoDB与REST相结合,利用MongoDB的高可靠性、高可用性以及高可扩展性的特点,以应对海量数据带来的挑战,利用REST的高灵活性和高可扩展性的特点,解决通航公司业务系统众多且业务变化频繁的问题。通过两者的结合,可以很好地满足通航产业快速发展的需求,为用户提供高效稳定的云服务。4)本文实现了MonogoDB-Hadoop机器学习平台。在大数据时代,海量数据往往蕴含着令人难以想象的价值,为了实现对数据的有效处理,将MongoDB擅长数据存储的优点和Hadoop擅长大数据分析的优点相结合,借助机器学习平台,为用户推荐热门航线,优化飞行航线等,提高系统的数据分析能力。更多还原

【Abstract】 In recent years,China’s general aviation has made a lot progress in scale and quantity. With further opening low-altitude airspace in 2015,general aviation industry will enter a high-speed development of the fast lane.Although our general aviation development has huge potential,the level of information is far behind developed countries.Data management of most general aviation still remain in the primitive stage of paper records.Although some general aviation corporations have their own information management system,the traditional dispersed data storage and isolated general aviation corporations result in that the data cannot be shared and stored effectively and that the reliability and security of the data cannot be guaranteed. What’s more, data mining cannot be used to extract the useful information from the stored data.The low information level and backward management have become the bottleneck of our country’s general aviation development.Faced with the informatization problems and actual demand of our general aviation industry, the paper combines MongoDB and the REST,using Hadoop data analysis platform, to build general aviation cloud data center.All the general aviation’s data will be centrally managed by cloud data center. Cloud data center will establish a communication channel within general aviation corporations and between companies and customers.Cloud data center will greatly improve flight efficiency,guarantee flight safety and improve the level of information.In this paper,main work and innovations are as follows:1) Analysis of cloud data center technologies. Depth study of the cloud computing architecture and core services, basic theories of NoSQL database, MongoDB and REST related contents and MapReduce programming model.2) Study the needs to cloud data center systematically, defined the function of the system. Have a deep analysis of the core business and determine the data processing sequences.3) This paper introduces the contents about the design and implementation of cloud data center of general aviation based on MongoDB and REST.Cloud data center uses document database MongoDB to store data which is one kind of NoSQL DBMS and uses REST interface to provide unified data access services for general aviation corporations’ business systems including PC and mobile platform.MongoDB with high reliability, high availability and scalability features, can easily cope with the challenge of mass data storage.With high flexibility and scalability characteristics, REST can solve the general aviation’s problems about numerous business systems and frequent business changes.Combined MongoDB and REST, the cloud data center can satisfy the needs of the rapid development of the general aviation industry and provide users with efficient and stable cloud services.4) This paper implements MongoDB-Hadoop machine learning platform.In big data era, massive data often contains unimaginable value.In order to realize efficient data processing, the system combines the MongoDB’s advantage of data store and Hadoop’s advantage of data analysis.With the help of the machine learning platform, the system can recommend popular routes and optimize flight path for the customers.So the platform can improve the system’s ability of data analysis.更多还原