Introduction to Apache SOLR: Lucene-based Scalable Cluster search Server

The Apache SOLR project is an open-source enterprise search Server Based on Apache Lucene, and version 1.3 has been released recently. Infoq interviewed the SOLR creator yonik Seeley and learned more about the new version and the functions provided by SOLR to end users.

Seeley first describes the target user: "search box, faceted browsing (navigation), or anyone who combines the two". The key features of SOLR include:

• Standard-based open interfaces-- The SOLR search server supports querying and retrieving results through XML, JSON, and HTTP.
• Easy to manage-- SOLR can be managed through HTML pages, server statistics are output in JMX, and SOLR configuration is completed in XML.
• Split View-- The search result is automatically classified.
• Highlight hit words-- The matched characters are automatically highlighted in the search results.
• Scalability-- Fast incremental update and snapshot distribution/replication to other servers.
• Flexible plug-in system-- New features can be conveniently added to the SOLR server in the form of plug-ins.

Seeley also talked about the main new features in this version:

• Distributed search-- Indexes can now be divided transparently into multiple parts. A single SOLR server supports multiple indexes based on various configurations and modes, and the main configuration can be changed without stopping the SOLR server.
• Extended query Functions-- Includes a new Java client (solrj) and a number of new features, such as directly configuring which documents are hit first, approximate hit, search expiration, record split time, and spell check for specific queries
• Enhanced Data Import tools-- Databases and other structured data sources can now be imported, mapped, and converted.
• More customizable extension points-- There is a new update processor chain that allows you to modify and redirect documents during query. A Search Component chain allows you to modify and add query results, user query analyzer, and plug-in functions.
• Performance Enhancement-- Significantly improves the indexing speed, binary response format, and quick query and deletion functions.

Detailed update logs can be obtained here.

Seeley talked about more details about SOLR's scalability, functionality, and Practicality:

SOLR has deployed millions of documents. With distributed search, SOLR should be able to process billions of documents.

SOLR is based on Lucene and has excellent full-text relevance. It can easily provide phrase closeness enhancement, recent document enhancement, editing enhancement, and customized scoring Mechanism Based on numeric value proprietary functions.

AOL is using SOLR to enhance its channel functions: Music, rugby, recipes, Reference Centers, real estate and cars all use this technology. SOLR's search function is also applied to Netflix, Zappos, gamespot, and Internet Archive. AndManyI can't reveal this to my major customers.

Regarding SOLR's future plan, Seeley mentions more scalability, more convenient configuration and management for large clusters, regional and real-time search and refactoring to use spring configuration plug-ins. Seeley also provides a mailing list where he discusses in detail the future of SOLR, especially the 2.0 plan.

View Original English text:Apache SOLR: extensible, clustered search server built on Lucene

Translator Cui KangHe is the translation editor of infoq Chinese site and is engaged in enterprise-level Web applications, focusing on performance optimization, Web technology, browsers and other fields.