Viewing 15 posts - 7,321 through 7,335 (of 7,602 total)
SQLKnowItAll (7/9/2012)
ScottPletcher (7/9/2012)
I know why the date was left out of the Sales table -- it can be derived.
What am I missing... How can it be derived?
Join...
July 9, 2012 at 2:17 pm
Gullimeel (7/9/2012)
Just a small thing.Why do you have a fill factor of 80 for increasing identity field?Shouldnt it be 100%?
I agree that the fillfactor should be carefully reviewed, but not...
July 9, 2012 at 2:14 pm
OmegaZero (7/6/2012)
I'm not 100% sure of how to handle the AccountSales table though - since if I did it by id the dates would not match up correctly if joined...
July 9, 2012 at 2:07 pm
Dev (7/9/2012)
ScottPletcher (7/9/2012)
Dev (7/9/2012)‘select only a very small % of the total rows’ is applicable to Clustered Index as well for seek operation else it will scan the cluster.
False, of...
July 9, 2012 at 10:29 am
Dev (7/9/2012)‘select only a very small % of the total rows’ is applicable to Clustered Index as well for seek operation else it will scan the cluster.
False, of course. ...
July 9, 2012 at 10:04 am
Dev (7/8/2012)
July 9, 2012 at 9:20 am
I suggest avoiding "WEEK" or anything that could or does rely on SQL settings such as @@DATEFIRST (I don't think WEEK actually does, but I think it does always use...
July 6, 2012 at 1:09 pm
case
when degree1id = "@Degree" then (select d.Name from AcademicExperienceTypes d where p.Degree1id=d.Id)
when degree2id = @Degree then (select d.Name from AcademicExperienceTypes d where p.Degree2id=d.Id)
when degree3id = @Degree then (select...
July 6, 2012 at 12:34 pm
case
when degree1id = "@Degree" then (select d.Name from AcademicExperienceTypes d where p.Degree1id=d.Id)
when degree2id = @Degree then (select d.Name from AcademicExperienceTypes d where p.Degree2id=d.Id)
when degree3id = @Degree then (select...
July 6, 2012 at 12:29 pm
The most important performance aspect of a table is determining the correct clustered index.
Absolutely, and what I want is one that is narrow - doesn't make nonclustered indexes larger than...
July 6, 2012 at 12:16 pm
GilaMonster (7/6/2012)
There's only one clustered index, multiple nonclustered indexes, clustered index to organise the table, nonclusters to support the queries.
Every column added to a query invalidates the existing covering index...
July 6, 2012 at 10:49 am
An integer column defined as identity (a very good and popular choice for clustered index) is 4 bytes.
It's certainly popular, but most often not good. The no-thought identity as...
July 6, 2012 at 8:40 am
CELKO (7/1/2012)[/bCode should be in Standard SQL as much as possible and not local dialect.
Given that this a SQL Server forum, SQL server dialect is fine. In particular,...
July 5, 2012 at 4:26 pm
Lynn Pettis (7/3/2012)
ScottPletcher (7/3/2012)
Also you have 2 very big indexes.One has 10 columns and other has 9 columns .Table just has 12 columns.
It's a pretty sure bet you have the...
July 3, 2012 at 3:45 pm
Also you have 2 very big indexes.One has 10 columns and other has 9 columns .Table just has 12 columns.
It's a pretty sure bet you have the wrong clustered index....
July 3, 2012 at 1:56 pm
Viewing 15 posts - 7,321 through 7,335 (of 7,602 total)