Best practices for search in SharePoint Server 2010

적용 대상: SharePoint Server 2010

마지막으로 수정된 항목: 2016-11-30

This article is one of a series of Best Practices articles for Microsoft SharePoint Server 2010. This article describes the best practices for enterprise search. Unless otherwise noted, this article applies to both SharePoint Server 2010 and Microsoft Search Server 2010. For more articles in the series, see 모범 사례(SharePoint Server 2010). For additional information about best practices for SharePoint Server 2010, see 모범 사례(SharePoint Server 2010) and 성능 및 용량 관리(SharePoint Server 2010).

In this article:

• Plan the deployment

• Start with a well-configured infrastructure

• Manage access by using Windows security groups or by using role claims for forms-based authentication or authentication using a Security Assertion Markup Language (SAML) security token

• Defragment the search database

• Monitor SQL Server latency

• Test the crawling and querying subsystems after you change any configuration or apply updates

• Review the antivirus policy

• Acknowledgements

1. Plan the deployment

• Plan the search topology and plan for capacity and performance. You must plan an appropriate topology for enterprise search to provide an optimal search experience. For information about how to plan a search topology, see 엔터프라이즈 검색용 토폴로지 계획(SharePoint Server 2010). For information about capacity planning for SharePoint Server 2010 search, see SharePoint Server 2010 Search의 성능 및 용량 요구 사항 예측. For information about SharePoint Server 2010, see 성능 및 용량 관리(SharePoint Server 2010).

• Plan for findability. For a search technology to be useful, users must be able to easily find what they are looking for. For a good discussion of findability, see Organizing Information in SharePoint Server 2010 (https://go.microsoft.com/fwlink/p/?LinkId=226827) by Microsoft MVP Bill English.

• Use managed properties and scopes to enhance the end-user search experience. Managed properties enable search administrators to create a one-to-many mapping of related properties to help users refine their search queries. For example, a search administrator can map crawled properties named "writer" and "author2" to the managed property "author" so that users who include the "author" property in their query also get search results for "writer" and "author2." For more information about managed properties, see 검색에 대한 메타데이터 속성 계획(SharePoint Server 2010). You can help users refine their search queries by using search scopes, which limit searches to a content subset, such as a specific site. For more information, see 검색 범위 계획(SharePoint Server 2010). If you use custom queries, make sure that the appropriate properties are set to be used in scopes. For more information, see 검색에 대한 메타데이터 속성 관리(SharePoint Server 2010)

• Plan for IFilters .The crawler uses IFilters to read individual file types when it crawls content. If you are currently using 32-bit IFilters in a Microsoft Office SharePoint Server 2007 farm, you must obtain 64-bit IFilters from the appropriate vendor for SharePoint Server 2010. For information about the 64-bit IFilters that are included in SharePoint Server 2010 by default, see File types and IFilters reference (SharePoint Server 2010).

• Create service level agreements. Ensure that service level agreements (SLAs) for content crawls are agreed to before deployment. For information about how to plan content crawls, see 크롤링 및 연결 계획(SharePoint Server 2010).

2. Start with a well-configured infrastructure

• Deploy two or more query components and crawl components for redundancy and increased availability. For more information, see SharePoint Server 2010용 검색 아키텍처: 모델. Increasing the number of query components and crawl components can improve performance. For more information, see SharePoint Server 2010 Search의 성능 및 용량 요구 사항 예측.

• Use at least two database servers to host the content databases and the databases for the Search service application (including crawl, property, and administrative databases). For more database recommendations, see 저장소 및 SQL Server 용량 계획 및 구성(SharePoint Server 2010) and SharePoint Server 2010 팜의 SQL Server 2008 모범 사례.

• Use a gigabit network for intra-farm connections.

3. Manage access by using Windows security groups or by using role claims for forms-based authentication or authentication using a Security Assertion Markup Language (SAML) security token

We recommend that you add users to Windows security groups or role claims instead of adding users to SharePoint groups for the following reasons:

• Because changes to Windows security groups do not directly affect the access control entries (ACEs) on SharePoint sites, you do not have to perform a new crawl when user accounts within those Windows security groups are changed.

• During the indexing process, the system stores the ACE of each user who was added to a SharePoint group instead of the ACE of the SharePoint group itself. This process supports approximately 1000 users per access control list (ACL), after which the "Parameter is incorrect" error causes crawling to fail.

For more information, see 보안 그룹 선택(SharePoint Server 2010) and 클레임 인증 구성(SharePoint Server 2010).

4. Defragment the search database

The search database contains metadata and ACLs of crawled content. Over a series of crawls, the search database can become fragmented. To improve performance of crawls and queries, periodically defragment the search database. For more information, see SharePoint Server 2010에 대한 데이터베이스 유지 관리.

5. Monitor SQL Server latency

Search is I/O intensive for SQL Server. For best search performance, we recommend that you follow the guidance for small, medium, and large farms that is provided in SharePoint Server 2010 Search의 성능 및 용량 요구 사항 예측.

6. Test the crawling and querying subsystems after you change any configuration or apply updates

We recommend that you test the crawling and querying functionality of the server farm after you make configuration changes or apply updates. An easy way to do this is to create a temporary content source that is used only for this purpose. To test, we recommend that you crawl ten items — for example .txt files on a file share — and then perform search queries for those files. Make sure that the test items are currently not in the index. It is helpful if they contain unique words that will be displayed at the top of the search results page when queried. After the test is complete, we recommend that you delete the content source that you created for this test. Doing this removes the items that you crawled from the index and those test items will not appear in search results after you are finished testing. For information about crawling content, see 크롤링 및 연결 계획(SharePoint Server 2010) and 크롤링 관리(SharePoint Server 2010).

7. Review the antivirus policy

When you use certain file-level antivirus software programs in SharePoint Server 2010, you should exclude certain folders from being scanned. If you do not exclude these folders, you can experience many unexpected issues. For more information, see the following article in the Microsoft Knowledge Base: 952167: Certain folders may have to be excluded from antivirus scanning when you use a file-level antivirus program in SharePoint (https://go.microsoft.com/fwlink/p/?LinkId=123963).

Acknowledgements

The SharePoint Server 2010 Content Publishing team thanks the following contributors to this article:

• Luca Bandinelli, Microsoft SharePoint Customer Advisory Team

• Dan Blood, Microsoft SharePoint Customer Advisory Team

• Sid Shah, Microsoft SharePoint Search Product Team

• James Sturms, Microsoft SharePoint Search Product Team