Nonclustered Columnstore Indexes provide a significant query performance improvement for relational data warehouses by grouping and storing data for each row and then join all the rows to complete the whole index.
When we use nonclustered columnstore indexes data will be grouped and stored by column. The main advantages we could get from column store indexes are batch execution instead of row by row execution. The absence of key column concept for nonclustered columnstore indexes makes it available for more than 16 columns. Nonclustered columnstore indexes works with the existing table partitioning. The maximum index size 900 byte doesn’t apply for nonclustered columnstore indexes. Although this might not be true for recently released SQL Server 2014 tables with nonclustered columnstore indexes cannot be updated
There are also some other restrictions such as it cannot have more than 1024 columns, cannot be clustered index, cannot be unique index, cannot be used for views or indexed views, cannot include sparse column, cannot be altered using ALTER INDEX, cannot have INCLUDE keyword and cannot have ASC OR DESC. Nonclustered columnstore indexes cannot be combined with features like page and row compression and vardecimal storage format (A columnstore index is already compressed in a different format.), replication, change tracking, change data capture and filestream.
Read more
http://technet.microsoft.com/en-us/library/gg492153.aspx
http://technet.microsoft.com/en-us/library/gg492088.aspx
