Viewing 15 posts - 1,141 through 1,155 (of 3,348 total)
Use the data manipulation tools in SSIS to remove columns you do not need, or to add extra columns with default values; then combine the results.
January 13, 2016 at 2:37 am
A normal view is like a macro definition in a third-generation language. It gets replaced by its definition before the query optimizer even sees it.
When you add an index to...
January 13, 2016 at 2:36 am
We know way too little about your tables and data, so we cannot answer your questions.
We need:
* The structure of all involved tables, posted as CREATE TABLE statements. You can...
January 13, 2016 at 2:28 am
The execution plan you describe is optimal for this query, you will not be able to speed it up.
For future maintenance, you can rewrite the query though. To this:
SELECT A.F_Name,...
January 13, 2016 at 1:59 am
If you insist on basing your ideas on relational theory on implementation choices made by Microsoft, then I guess we'll have to agree to disagree on this.
January 12, 2016 at 5:06 pm
I don't understand most of the terms you use about your business, but based on the sample data I think your table design is not normalized.
You'll want to have two...
January 12, 2016 at 5:02 pm
You'll have to test to be sure, but I expect the answer to be no.
Fragmentation of indexes mostly affects scan performance (which includes seeks that are not for singleton rows!).
Modification...
January 12, 2016 at 4:57 pm
Sergiy (1/11/2016)
No, I'm not confusing anything.DATETIMEOFFSET is indeed made up from the different values having totally different meaning.
If you think that datetimeoffset violates relational theory, then you should not take...
January 12, 2016 at 3:51 pm
ps_vbdev (1/12/2016)
January 12, 2016 at 3:33 pm
Eric, I guess I didn't make my point well enough. What I was trying to stress is the importance of normalizing data as early as possible. If you can normalize...
January 11, 2016 at 3:38 pm
I like to compare index structures to books. The comparison does not hold all the way through, but it does well for most of the basics.
A nonclustered index is like...
January 11, 2016 at 3:26 pm
Grant Fritchey (1/11/2016)
You can, if necessary, apply a clustered index to the table and then drop that index.
In most cases, the better idea is to stop halfway through that procedure.
(I.e.,...
January 11, 2016 at 3:19 pm
Rankerg (1/11/2016)
So creating a store procedure and like to pass table name as parameter?
No.
January 11, 2016 at 3:14 pm
Sergiy (1/10/2016)
It contains 2 different values in 1 placeholder, which is prohibited by Relational Theory.
Values with delimiters ("T" in this case) belong to the...
January 11, 2016 at 3:13 pm
Viewing 15 posts - 1,141 through 1,155 (of 3,348 total)