Viewing 15 posts - 5,731 through 5,745 (of 22,211 total)
I'm with @chrism-2. Let's see the execution plans so we can understand what each server is doing to resolve these queries.
How are your statistics being maintained on each server?
August 11, 2015 at 3:18 am
To read the entire table... probably not. Getting rid of the FOR XML will be faster. Other than that, faster and more disks, faster and more controllers. Since you're reading...
August 10, 2015 at 3:41 pm
Yeah, don't spend your money getting certified when you don't have any experience to back it up. Certification is highly unlikely to get you a job. You'd be better off...
August 10, 2015 at 11:25 am
sgmunson (8/10/2015)
Grant Fritchey (8/10/2015)
August 10, 2015 at 9:01 am
A lot of the functionality that was introduced in 2012, Availability Groups, ColumnStore indexes, others, are much improved in 2014. Also, the 2014 cardinality estimator is completely new and improved,...
August 10, 2015 at 8:58 am
SQL Server 2016 is not actually released yet, so I wouldn't recommend it for a production environment (legally, unless you're part of Microsoft's TAP, you can't). However, SQL Server 2014...
August 10, 2015 at 8:46 am
Might be a good idea to put an index on your temporary table and ensure that the statistics are up to date. This query is the most costly:
SELECT TOP 1
...
August 10, 2015 at 8:44 am
Sean Lange (8/10/2015)
mohan_padekal (8/10/2015)
Here is the Execution PlanHow about as an attachment instead? I doubt anybody is going to waste their time piecing that back together into something usable.
+100
Please, attach...
August 10, 2015 at 7:45 am
Back to comparing the execution plans to understand where the differences occur.
August 10, 2015 at 7:44 am
You're going to spend a lot more time trying to create this magic query for each table that will dynamically allow INSERT/UPDATE/DELETE/SELECT than it will take you to just create...
August 10, 2015 at 5:11 am
Can you post the execution plan? On an initial read through the procedure I didn't spot any immediate issues, so it's a question of understanding how each of those actions...
August 10, 2015 at 5:04 am
With the data stored in a denormalized fashion like that, you've really got two choices. The better option, break up the name into multiple columns. However, that's going to be...
August 10, 2015 at 5:01 am
UserRR (8/10/2015)
I actually read that whitepaper and got that trace flag from there.
I understand some queries might degrade, but how can the query perform 15x better with no parallelism?
Understand...
August 10, 2015 at 4:55 am
August 10, 2015 at 4:48 am
I can't replicate it here. Check your SET ROWCOUNT in the properties. Maybe that's set to 11.
August 9, 2015 at 5:46 am
Viewing 15 posts - 5,731 through 5,745 (of 22,211 total)