In this case, this particular report will look at the orders by
customer location. Also, in the Adventure Works sample, the end customer is known
only for Internet orders. Therefore, the Internet Sales Amount measure is brought
over from the appropriate measure group and dropped on the design surface. A
value now appears immediately; this is the sum of all Internet Sales for all time, all
products, all customers, and so forth.
Since this will be a report about customers and their sales, it might be tempting
to simply drag the entire Customer Geography hierarchy onto the design surface.
Certainly this will work, but you should understand the impact: the query will now
go down to the lowest level of the hierarchy, which in this case is the individual
customer. That means that the query in the background will be retrieving a dataset
that is potentially very large, and both time and memory constraints may enter into
the picture. Figure 10-5 actually shows the entire Customer Geography hierarchy in
the design surface. For small cubes this may not be an issue, but in the real world
you may find that going down to the leaf level can lead to issues, especially when
there are multiple dimensions on the report, as there will be in a moment.
By simply having the Customer Geography hierarchy you know that there is
only one dimension in the report at the moment. This might work in some cases, but
many reports also include time, so for this report the Date.
Pages:
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412