SQL Server 2008 – Filtered Index
Granted that SQL 2012 is already out, there is actually a feature inside SQL 2008 which is especially useful for some form of slice and dice when you have too much data.
This feature is called Filtered Index. You can find more information about it at http://technet.microsoft.com/en-us/library/cc280372.aspx.
Generally the purpose is to improve query performance by storing indexes only for a subset of the data. Additionally it reduces costs and storage incurred with index maintenance.
The syntax to create one is as follows
create unique nonclustered index [IndexName] on dbo.[TableName]([Fields])