You are joining here the forcast on chain and location. Do you need to join on both conditions? In the visualisation tool there may be a setting to treat this...

It fails because the underlying table was created with ANSI_NULLS set to off.

Old post, but I ran into this issue recently.

I get an error when trying to create a 'filtered' non clustered index only. A non filtered variety works just fine. Seems...

This is an excellent diagram. I saved a copy as well.

The problem starts with management. We live in a time when people have short attention spans and want to see results now. Imagine telling a top level manager that what...

A report user would still want to see not only what items sold, but what did not sell all in the same view. So doing a right join (sales events...

As far as Primary Key (without Clustered Index) is concerned you MAY have it on a GUID or an Identity column (mostly preferred). But if you ask Primary Key...

I would not advise a GUID as a primary key. They are 16bytes in size last time I checked, have no fixed order, and a int or big int could...

One approach you can take is create your own surrogate key for all record and keep the customer key as its own column.

"Policy Number would make a decent clustered index...

With OLAP your dimension tables are typically wide and short compared to your fact tables (tall and narrow). Typically we add everything to the dimension if the attribute is a...

You can use a case statement  along with the functions Grant alludes to determine the column  datatypes. I hope you only have to do this once per column / per...

Data type disparity on join conditions etc is not really an after thought I've seen over time, its more like a 'no thought.' You bring up an excellent point that...

Dont forget any secondary replicas if you are using High Availability.

• This reply was modified 4 years, 10 months ago by MMartin1.

It would default to varchar(30) , but still to wide for a...

I dread the day that machines will not need to depend on us.