Sunday, November 7, 2010

SharePoint 2010 Farm Topologies

Normally a SharePoint Farm topology vary on number of factors.

1) Number of users.
2) Redundancy requirements.
3) Service applications being used.

Looking back at SharePoint 2007, it was SSP (aka Shared Service Provider) we had. But with the new service applications model in SP 2010 the users have greater flexibility in deployments.

Given below are the different Farm Topologies used in SP 2010 Deployments.

Single Server Farm
This is ideal for a user base of less than 100. This farm consists of a single server catering the requirement of the Web Front End (WFE), Database Server and the Application Server.

image

Two Server Farm
This farm is equipped with one Database Server and one Webserver that performs all the application services. For high availability a clustered or mirrored database server is recommended. This farm can cater the requirement of 100 to 10,000 user base.
image

Two-Tier Small Farm
This environment is adequate for a user base of 10,000 to 20,000 with low service usage. Two Web Servers, one Web Server performing the tasks of the Application Server and the Database Server (clustered or mirrored).
image

Three-Tier Small Farm
Three-Tier small farm is much more similar to a Two-Tier Small Farm. A dedicated Application Server is added to cater the requirements of the Services. To improve the performance of the Search Service a the Search Database can be moved to a dedicated Database server. This Farm Topology is ideal for a solution with a Search Database containing nearly 10 million items which can be considered as a large Search Database. Though its optional, it’s a good practice to have a Database Cluster or a mirror.
image

Medium Farm
This is a Three-Tier Farm. The general rule in planning is to have 10,000 users for a WFE Server. In the First Tier we can utilize two or more Web Servers. In the Second Tier there are two servers dedicated to crawling contents and serving search queries and one or more servers for all other application services. In the Third Tier there are dedicated servers for Search Databases and one or more servers can be used for all other SharePoint Databases.
image
The middle tier can be expanded to to handle the growth by adding more servers to cater the requirement of application services.
*optional servers are in gray color.

Large Farm
A Large Farm is built on the Server Group concept as it used in the Medium Farm. Additionally a Large Farm could consist of dedicated servers for Sandboxed code.
image

It's always recommended to use the concept of “server groups” to group services with similar performance characteristics together onto a single server and then add servers based on the needs of those particular services.

9 comments:

asif said...

Thank you so much for sharing this nice post about Farm Topologies. This will help me a lot in my assignment:)
Free Xavor’s SharePoint 2010 tools

sumeet said...

What are web, application and db servers? each does what?

Prabath said...

@Sumeet, in this scenario Web Server is IIS (Internet Information Services), Application Server is where we host the services (ex: Search Service, crawling service, Performance point) and Database Server is Microsoft SQL Server.
I hope this helps.

Raghav said...

Nice post. Got an idea about farm topology. Thanks

Đỗ Quốc Hùng said...


Good article, i would like as something, can you help me that blow question.
1. how many topologies
2. each topology list show advance and disadvance
3. each topology how many server
4. server role in topology đó (front-end, back-end, database, services ...)
5. Plan how the solution will respond to increased workloads, changing user requirements, and server failures
6. Ensure that your solution is scalable, flexible, and resilient to hardware failures

Đỗ Quốc Hùng said...


Good article, i would like as something, can you help me that blow question.
1. how many topologies
2. each topology list show advance and disadvance
3. each topology how many server
4. server role in topology đó (front-end, back-end, database, services ...)
5. Plan how the solution will respond to increased workloads, changing user requirements, and server failures
6. Ensure that your solution is scalable, flexible, and resilient to hardware failures

Roberto said...
This comment has been removed by the author.
Đỗ Quốc Hùng said...

hi Roberto. thanks for your answer. after some months, i clear my problem. thanks so much.

Kapil Garg said...

Nice post bro..