Font Size: a A A

Research On Migration Algorithm From Traditional Relational Database To Non Relational Database

Posted on:2018-12-19Degree:MasterType:Thesis
Country:ChinaCandidate:C H SongFull Text:PDF
GTID:2348330536474666Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
In recent years,with the application of Internet of things and the proposing of the concept of “Internet Plus”,the amount of data in the Internet showing a trend of exponential growth then enter the era of big data from the Information age.The data in Big Data Time not only has a rapid growth but also includes the change of structure,distribution,value density,dimensions and so on,which needs non relationanal database that with the characteristics of semi structured and unstructured data store,high scalability and high read and write and distributed computing modle to meet the requrements in parallel processing on high scalability.However,the traditional relational database can not meet these requirements,which has accumulated a large amount of data because of its extensive using in this years.Therefore,in order to make full use of these data to continue to provide services for users and make more values for enterprises need to design an automated or semi automated migration tool to complete the change from traditional relational database to relational database.In addition,due to the use of the interface of non relational data and traditional relational data has an essential difference,it is necessary to encapsulate the native interface of the non relational database by implement a SQL Middleware that can complete this change to make the original business applications can adapt to the new data storage model and to promote the smooth transition of traditional relational database to non relational database.In this paper,through the research of the database storage structure,the big data correlation technology and the operation interface of the non relational database HBase,the database migration tool is designed and implemented,and the query interface of HBase is encapsulated.Firstly,the related technologies of data storage and migration are studied,in-depth analysis of the storage principle of traditional relational database and design method,HBase database architecture and storage mode,Hadoop ecological system and the related data migration model design and technology.Secondly,according to the deep analysis of the metadata of the traditionalrelational database,this paper proposes and designs the algorithm of table schema ad data's migration from MySQL to HBase.Then,the native API of HBase is analyzed in depth,and SQL Parser and Filter are used to filter and combine the operation of traditional relational database.Finally,through the design of experiments carried out performance testing,analysis results show that the migration tool and operation middleware in this paper can effectively complete the migration work of historical data and use the process in the future to meet the query requirements,provide a guarantee for the transition from relational database to non relational data.
Keywords/Search Tags:Relational database, Non-relational database, Data transfer, Multi fork tree structure model, Data query
PDF Full Text Request
Related items