For upp to at least CMS 6 this isn’t EPiServers best functionality – please don’t use it. But if you have...
But if you have
The search functionality in EPiServer are based on several ways of searching for the data
- Text search in the database to find content in EPiServer
- File search with Microsoft
Indexing Service for searching files in the
native file system
- Lucene for search files in the versioned file system
EPiServer Full Text Search in EPiServer Framework 6.1
The EPiServer Full text search (FTS) is a new EPiServer property search engine that is able to search for properties in a lot of database content. It's well written with a WCF REST Service that can be stored on a separate machine.
It’s great when you want to search for text in properties, and can list pages that include the word you search for (great for the editorial view). But when you want to implement good search functionality for your web users it misses the most important functionality as:
1) Basing search result on relevance of the content
2) Don't index web sites as a web crawler
EPiServer FTS ClientEPiServer FTS ServiceSolr with EPiServer Full Text Search
There exist several module projects to EPiServer that uses
Solr that index a web site with a web crawler and order the search result with relevance to the content. I haven't tied it yet but I think this is the right way of getting a good and cheap search for your site.
Other search engines
Gogle GSA/MiniAn excellent search functionality easy to implement, and delivers what I expect it to
deliver.
Microsoft Search ServerA Great search functionality that is easy to implement with .NET API that is easy to
understand.
EuroLing SiteSeekerOne of the world’s greatest support organizations, easy to implement if you accept
100% standard EPiServer search. Otherwise it has a hard to understand API against
WebServices. A mayor design issue is that the search engine is not using UNICODE
so it’s not able to handle Chinese, Korean and so on.
EasySearchA commercial search plug-in for EPiServer based on Lucene. This is a property search engine that only searches directly in EPiServer properties so you shouldn’t expect as it work as expected as search functionality for external users. It makes searches quick with a low CPU utilization of the web server.
The latest version I worked with was released 2009 and doesn’t support a load balanced environment, because the Lucene index files uses locks and can’t be located on a network drive. And you have to invent it by yourself (takes about 40 hours).
[SolR]
Solr is an abstraction layer for Lucene written in Java. The Java server is easy to install and use on a windows server (the most common scenario is that installed on the public web front end.
iTera consulting (in swedish) has developed a EPiServer property search implementation that’s work very well with an easy API, the drawbacks if that you are dependent of iTera if you have a low knowledge of Java and SolR in your company.