Because the search engine features in the portal community to enhance the user experience has focused on the portal community involved in a large number of search engine requirements, there are currently in the implementation of the search engine is a centralized solution to choose:
1. Based on Lucene self-encapsulation to achieve in-station search. Workload and scalability are large, not used.
2. Call Google, Baidu's API to implement the site search. With third-party search engine binding too dead, unable to meet the late business expansion needs, temporarily not adopted.
3. Based on Compass+lucene to achieve in-station search. Suitable for indexing database-driven application data, especially in lieu of traditional like '%expression% ' to index fields such as varchar or CLOB, it is a worthwhile scheme to implement in-site search. But in the distributed processing, the interface package still need to carry out some degree of encapsulation, temporarily do not adopt.
4. In-site search based on SOLR. Encapsulation and extensibility are better, providing a more complete solution, so in the portal community to adopt this scheme, later joined the Compass program.
1. about SOLR
SOLR is a lucene-based Java search engine server. SOLR provides level search, hit highlighting, and supports multiple output formats (including XML/XSLT and JSON formats). It is easy to install and configure, and comes with an HTTP-based management interface. SOLR has been used in a number of large sites, more mature and stable. SOLR has packaged and expanded Lucene, so Solr basically follows the terms of Lucene. More importantly, theindex created by SOLR is fully compatible with the Lucene search engine library. by properly configuring SOLR, some situations may require coding,and SOLR can read and use indexes built into other Lucene applications. In addition, many Lucene tools (such as Nutch, Luke) can also use the index created by SOLR.
2. Tomcat under SOLR installation configuration
Since SOLR is based on Java development, SOLR is well-deployed in both Windows and Linux, but since SOLR provides some shell scripts for testing and managing and maintaining more convenient Therefore, it is recommended to install on Linux during the production deployment, which can be used in Windows during testing.
Construction and use of SOLR under Linux (recommended jdk1.8 above)