large volume of data

Last Post 23 Feb 2007 04:49 AM by JHunter. 0 Replies.
AddThis - Bookmarking and Sharing Button Printer Friendly
  •  
  •  
  •  
  •  
  •  
Sort:
PrevPrev NextNext
You are not authorized to post a reply.
Author Messages
JHunter
New Member
New Member

--
23 Feb 2007 04:49 AM
Firstly I'd be looking to redesign the query, 250000 rows being returned to a client is simply unusable (unless the client is performing significant client side analytical work).

The approach you're looking for is know as paging, using the Row_Number() function with in a stored procedure which takes the start and end rows for a specific page is a good start.

If you require a client session to run in isolation of changes, look into using snapshot isolation - but large data sets will make this perform really badly!

Partitions will certainly speed up your queries (if the partitioning is designed correctly), but are not designed for "paging".

Jamie
You are not authorized to post a reply.

Acceptable Use Policy