Abstract: With the rapid growth of network information resources, traditional retrieval system has been difficult to provide efficient and reliable services to the mass data. In response to this situation, this paper designs a distributed full-text retrieval system based on Solr. The system uses a Web crawler to collect information which is stored as text files. Then the system creates indexes in parallel on multiple computers through Solr index module. It turns out that the design improves the indexing speed effectively. The system improves the retrieval performance by applying Zookeeper management and distributed design in search module. Finally a user-friendly interface is designed. Currently, the system can operate millions of data stably and has a strong practical value.

Key words: full-text search, Solr, distribution, Zookeeper