Design And Implementation Of University Library Cloud Platform Based On Hadoop

Currently,"Library Alliance" is an important trend in the development of university libraries.For the purpose of sharing and reciprocity,through the sharing of resources such as books and electronic journals,the utilization of library resources has been greatly improved.However,the construction of "library alliance" in China is still in the primary stage.Due to the administrative management system and other reasons,many university libraries in China are independent of each other.And the phenomenon of heterogeneous data and "information island" is widespread.Therefore,if we can maintain an efficient information sharing platform based on the preservation of the original systems of each library,it will have great practical significance for promoting inter-library cooperation and resource sharing among university libraries.Based on this purpose,this paper designs and implements a university library cloud platform.It integrates and shares data from multiple university libraries through the research and application of distributed technologies such as Hadoop.Through the integration and display of library information,the university library is assisted to carry out better information interaction,and then promote the cooperation and sharing of university libraries.The main work of this paper is as follows:(1)Build a data integration center for the campus library.Aiming at the problem of data heterogeneity among different libraries,a data integration scheme based on Flume is proposed.This scheme can make the data extraction,cleaning,supplementation and standardization operations configurable through the rewritten source and sink units.Through this scheme,the heterogeneous data integration function in the library is realized.At the same time,aiming at the synchronization problem between original data and integrated data,a trigger-based data synchronization scheme is designed and implemented to ensure the effectiveness of integrated data.(2)Building a data integration center for multi-school libraries.Aiming at the massive and scattered data of multiple university libraries,a data integration system based on Hadoop is designed and implemented through the research and application of distributed technology.The system realizes the functions of data collection,processing and storage among multiple university libraries through the organic combination of distributed technologies such as Flume,Kafka,Storm,HDFS and Hbase.(3)Construct a visual application of the library cloud platform.Aiming at the effective utilization of multiple library data in the cloud platform,the analysis and query function of integrated data is designed and realized in detail through MapReduce and Hbase.And the visualization application of the cloud platform is realized based on the result data,which provides intuitive service for information interaction between university libraries.Through the final example operation,the university library cloud platform proposed in this paper can realize the integration and synchronization of data from multiple university libraries.It can effectively promote the information exchange and resource sharing among university libraries,and meet the actual needs at the current stage,and has strong practical application value.