Based On The Distributed Real-time Solr Full-text Retrieval System Design And Implementation

With the rapid development of information technology, more and more enterprises set up their own information platform or web site. When the amount of data within the enterprise is growing, use the database search function will seriously affect the search efficiency, to use the database to provide search capabilities to achieve search engine (Baidu, Google) retrieval efficiency is impossible. Solving this problem need to use the full-text retrieval technology which used in search engine, how to embed the full-text retrieval application in the enterprise IT system, and provide efficient retrieval service is the main problems that many enterprises need to solve. Enterprises often need system have strong performance and scalability, according to the enterprises require a higher real-time information characteristic, therefore a distributed real-time full text retrieval system can solve the above problems.This paper first introduces the distributed and full-text retrieval system, and detail description the core technology of the full-text retrieval and Lucene (a sub-project of Apache Software Foundation), Lucene is an open source full-text retrieval engine toolkit. In-depth understanding of Lucene, and analyzing its related projects, finally chooses to use its subprojects Solr as the core technology of the development of distributed real-time text retrieval system. Solr is an enterprise search application server, based on Lucene, developed by the Apache Software Foundation, which provides the client interface can easy to implement distributed applications, as an open-source projects and its development is for enterprise application, so it is very suitable for enterprise use.In this paper Solr’s architecture and features are discussed in detail, and highlights the build of the Solr server. In-depth study on Solr and Solrj (Solr client programming interface), detailed description the design of distributed real-time full text retrieval system, and puts forward a simple model of a distributed full-text retrieval system. Finally, the distributed real-time full text retrieval function integrates in a specific project, and does a simple performance testing on Solr retrieval server.