Frederick Staats

I'm worried about the scaling of paging very large result sets with DLINQ current implementation of Skip() and Take(). The internal mapping of using the ROW_COUNT() function with a large (millions of rows) result set will eat huge amounts of resources. There is much more efficient way to do paging with a two step query and SET ROWCOUNT (for an example see http://www.4guysfromrolla.com/webtech/042606-1.shtml). Has something like this been considered as an alternative implemenation of Skip() and Take() in DLINQ.


Re: LINQ Project General Alternate implemenation of Skip() and Take() on DLINQ

Matt Warren - MSFT

Frederick, thanks for pointing this out. We'll take it into consideration.




Re: LINQ Project General Alternate implemenation of Skip() and Take() on DLINQ

Frederick Staats

Thanks. This would also have the advantage that you could use a single highly performant solution for Skip() and Take() for both SQL SERVER 2005 and SQL SERVER 2000 instead of mapping to different solutions for each.