May 22, 2020 at 2:41 pm
To be clear, the link you posted from stack overflow cites TF 4199 and I'm not talking about that. I'm talking about TF 9481, which fixed my issues when, like you, I migrated from 2012 to 2016.
--Jeff Moden
Change is inevitable... Change for the better is not.
May 22, 2020 at 9:29 pm
In my opinion it is the same as "Legacy Cardinality Estimator" ON at database level.
If yes, I mention this on the very first page.
Martin
May 23, 2020 at 6:33 am
But is it is written on the very first page I do have it switched on - since the migration, I now about this switch. And I have still some problems, but fewer than without it.
On Database level.
May 23, 2020 at 7:13 pm
Doing things at the database level isn't necessarily doing the same as doing something at the server level. I strongly recommend that you try TF 9481 at the server level and see if that fixes things for you. It did for me. YMMV.
The other choice is to go fix all the queries that still have the problem.
--Jeff Moden
Change is inevitable... Change for the better is not.
May 26, 2020 at 12:28 pm
Sometimes the ultimate fix is a rewrite of the queries to a more optimal pattern.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
May 26, 2020 at 1:53 pm
Sometimes the ultimate fix is a rewrite of the queries to a more optimal pattern.
Totally agreed... sometimes, it's the only fix.
--Jeff Moden
Change is inevitable... Change for the better is not.
May 26, 2020 at 5:47 pm
Today I have a peformance issue where setting legacy cardinality estimator ON helps.
From 100s to 1 sec. Really interesting.
May 27, 2020 at 11:13 am
Today I have a peformance issue where setting legacy cardinality estimator ON helps.
From 100s to 1 sec. Really interesting.
If you have individual queries, and you can edit them, it is possible to add a hint to use the legacy estimator. I shy away from recommending this stuff because usually, the query needs editing anyway, so if we can modify the code, why not try to generally improve it rather than prop it up with a hint. However, it's a good thing to know.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
May 27, 2020 at 1:45 pm
Today I have a peformance issue where setting legacy cardinality estimator ON helps.
From 100s to 1 sec. Really interesting.
Heh... that's not just "help"... that's a freakin' miracle and we found enough of those where we just said "screw it" and set the whole server back to using the legacy Cardinal Estimator.
To be honest, I've not actually seen anywhere where the new Cardinality Estimator has actually made anything perform better. As I've been known to say, "Change is inevitable... change for the better is not".
--Jeff Moden
Change is inevitable... Change for the better is not.
Viewing 10 posts - 16 through 24 (of 24 total)
You must be logged in to reply to this topic. Login to reply