Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
Ali Chadorbaf
Starting Member
44 Posts |
 Posted - 2007-04-11 : 12:52:48
|
| Hi,I have a dimension table with over 7 million rows and 30 columns in each row. Only a few of columns are interested for most of the cubes, reports, and ad hoc queries. I am researching the techniques to improve query (and updating) performance on this dimension. I know logical partitioning and indexing are the most important. In addition, does splitting this large dimension to two mini dimensions not in the number of rows but in the number of columns help? I mean one mini dimension with the first few (and more interested) columns and another mini dimension with the rest (and less interested) columns. Does this improve the query performance?I’d appreciate to share your idea on this. |
|
|
|
|
|
Multimillion-row dimentions - performance techniqu