Vision

Enterprise Search in SharePoint 2010: brief discussion

Written by: Margriet Bruggeman, Nikander Bruggeman.

April 30, 2010

Enterprise Search in SharePoint 2010: brief discussion

Although the biggest change in the SharePoint search arena, as far as we're concerned, consists of the addition of Fast Search to SharePoint (see section "Fast Search Server 2010 for SharePoint"), there have been additions to the existing enterprise search features of MS Search as well. In this section, we will quickly touch on them.

You can now perform fuzzy name searches. If you search for "Nik", you may find "Nick", and "Nico" as well. Search center now allows you to use the Boolean operators AND, OR, and NOT. Search queries support wildcards at the end of a search term, for instance, you can search for "la*" and find results such as "large" and "laugh". You can also use Windows Explorer to search for documents in SharePoint sites. Search results can be navigated via a refinement panel (also known as drill-down or faceted search), that allow you to refine search results based on categories such as author, data, and document type. Search result summaries have been improved and include additional information such as contact persons, rating, and document tags. Search Center provides suggestions while you type a query. Site administrators can manage a central list of taxonomy keywords that are assigned to a specific column. Document owners can take these predefined and pre-associated taxonomy keywords and use them when definining document metadata, which makes a document easier to find. Searching for your own name yields additional information, such as the number of times your profile has been viewed in a given period, or keywords that have been used to find you. Probabalistic ranking has been improved in several ways: the click-through history has become an important ranking factor, if a specific document has been visited more than others it's ranking is promoted. Inferred metadata influences ranking as well, if a document is missing certain pieces of metadata (such as: it's author), the crawling process will try to infer the missing metadata in an intelligent way from the contents of a document. Finally, social distance has become an important ranking factor in people search. Search results of graphics in Image libraries now also include metadata such as it's size and last modification date. The current document indexing limit lies around 50 million documents. Enterprise Search is shipped with a connector for crawling Exchange 2010 public folders, and there is a framework for creating custom connectors too. Finally, topology management has improved as well. In SharePoint 2010, it is easy to increase the number of query servers and/or index servers. You can also configure a primary index server and a secondary, redundant index server to avoid system down time.

« back to overview page