Dimension Attribute/Join question
- From: leomepham@xxxxxxxxxxx
- Date: 18 Apr 2007 10:54:55 -0700
Hi, I'm having a problem and I wonder if I could get some ideas from
the crowd. I'm working with a database table that looks something like
this:
Event_Id - Event_SubId - Type - String - Value
1 - 1 - SalesMan - Bob - NULL
1 - 2 - Sales - NULL - 12
1 - 3 - Refunds - NULL - 6
2 - 1 - Salesman - Jim - NULL
2 - 2 - Sales - NULL - 16
2 - 3 - Refunds - NULL - 8
3 - 1 - Salesman - Fred - NULL
3 - 2 - Sales - NULL - 11
3 - 3 - Refunds - NULL - 5
Now designing the cube so sales & refunds are aggregated is no
problem, but assigning the correct salesman is tricky. My current
approach has been to create a parallel dimension using a pseudo-key
of the Event_Id and Event_SubId joined, then use that to put the
salesman against his sales & refunds. This works, but the problem with
this approach is it's very slow to process.
The other problem is that the 'Salesman' type object is not guaranteed
to be the first item listed, there are items that could apppear above
it. However, in each instance of Event_Id & Event_SubId it will only
appear once.
I'd be grateful for any suggestions or pointers.
Many thanks - Meff.
.
- Follow-Ups:
- Re: Dimension Attribute/Join question
- From: leomepham
- Re: Dimension Attribute/Join question
- From: leomepham
- Re: Dimension Attribute/Join question
- Prev by Date: Re: Inc Update of Dimensions using SSIS
- Next by Date: Re: Dimension Attribute/Join question
- Previous by thread: Inc Update of Dimensions using SSIS
- Next by thread: Re: Dimension Attribute/Join question
- Index(es):
Relevant Pages
|