Viewing 15 posts - 766 through 780 (of 1,353 total)
Is this a theoritical question? It seems odd that the same field would reference two tables, which means that the same data point has to be true in two different...
February 13, 2014 at 6:51 am
No because I already stated that both Husband and Wife are keys (i.e. candidate keys). Enforcing all the desired keys of a table is of course extremely important
The second...
February 12, 2014 at 11:15 am
My favourite example, a table of Marriages with exactly two attributes: Husband and Wife. Both attributes are candidate keys because clearly we don't want to allow the same person to...
February 12, 2014 at 9:11 am
What exactly is wrong with the Excel Destination in SSIS?
Although SSIS does right fairly well to excel files, I find I have to employ inelegant "tricks" when it...
February 11, 2014 at 7:25 am
I have not found what I consider a good MDX book that seems to focus on writing MDX queries for SSAS. Even one closely titled to that, something like...
February 6, 2014 at 7:08 am
You don't give enough information to give exact advice. However, I don't see the words "grain" or "level of granularity" in your question. A fact table is an...
February 4, 2014 at 11:32 am
I'm not sure I follow this completely, but my initial inclination would be to make this a single dimension with the initial heirachical attribute having as its data either Product...
January 24, 2014 at 7:25 am
Just because there are multiple paths doesn't mean that they are all equally valid. Just as in other things there are accepted practices that exist for good reason, so...
January 20, 2014 at 8:26 am
foreign keys add unncessary overhead to an ETL process.
This is true on the face of it, although whether that overhead is significant depends on the number of rows...
January 17, 2014 at 12:14 pm
I don't ever recommend dropping and recreating the warehouse from scratch, although I've seen that as a solution from some consultants. It's faster to develop but ultimately (and as...
January 7, 2014 at 8:16 am
I did mean to add the word difficult. Odd that I dropped an entire word. Letters I understand... I have altered the previous post.
December 6, 2013 at 9:36 am
If the only requirement is to run SSRS reports, the operational databases shouldn't be used and the data can be one day old, I would report off a restored backup...
December 6, 2013 at 8:05 am
Great article. Lots to think about. Any thought to testing deleting the rows to be updated so that everything is then an insert. I'm not necessarily keen...
October 31, 2013 at 9:06 am
In a few weeks I'll attend an analytics seminar where Kimball will deliver the keynotes.
While a signed book would be cool, the seminar would be the part I'd be most...
October 18, 2013 at 9:06 am
Don't worry that you're a long way. These take time. BI is a good path to go down. You should find it worth it.
October 16, 2013 at 8:39 am
Viewing 15 posts - 766 through 780 (of 1,353 total)