|
Home > Archive > MS SQL Server OLAP > November 2005 > Linked Dimensions
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
|
|
| Ron Van Zanten 2005-11-22, 8:24 pm |
| I have a best practices question. I am playing around with Linked dimensions.
My thinking is that we could develop a cube that has all of the common
dimensions in our DW (time,account,person
,etc...) in a cube and have linked
dimensions to the other cubes that need those demensions. My thinking is that
this would allow us to maintain only one dimenision and if needed when we
modify that dimension it would propogate through all of our cubes. This would
also allow for a consistant look and feel for the cubes that have common
dimensions.
However, it seems to be kinda of a struggle to keep consistancy with the
data source views for each of the "client" cubes. Also, I am concerned about
other unknown issues that could come up.
What does everybody think? Bad idea?
Ron
| |
| Darren Gosbell 2005-11-22, 8:24 pm |
| I thing a more typical usage pattern would be to have one cube, with all
the common dimensions in it and multiple measure groups. The on top of
that to provide easier browsing you would have multiple perspectives.
--
Regards
Darren Gosbell [MCSD]
Blog: http://www.geekswithblogs.net/darrengosbell
In article <C4D9A3BD-546B-4006-93BE- 87EC66B6521E@microso
ft.com>,
RonVanZanten@discuss
ions.microsoft.com says...
> I have a best practices question. I am playing around with Linked dimensions.
> My thinking is that we could develop a cube that has all of the common
> dimensions in our DW (time,account,person
,etc...) in a cube and have linked
> dimensions to the other cubes that need those demensions. My thinking is that
> this would allow us to maintain only one dimenision and if needed when we
> modify that dimension it would propogate through all of our cubes. This would
> also allow for a consistant look and feel for the cubes that have common
> dimensions.
>
> However, it seems to be kinda of a struggle to keep consistancy with the
> data source views for each of the "client" cubes. Also, I am concerned about
> other unknown issues that could come up.
> What does everybody think? Bad idea?
>
> Ron
>
|
|
|
|
|