September 13, 2023 at 5:09 pm
BTW, does anyone know if hardware (ISCSI controllers, CPU, memory) resolves the issues with 2022?
I know. Hardware shouldn't be required to fix bad code... but sometimes poor performance can be due to the wrong hardware being used from the get-go.
Quite certain that the chances of fixing the 2022 issues using hardware are as slim as the chances of fixing a clockwork by poking a needle into it whilst blindfolded.
😎
September 13, 2023 at 5:46 pm
Jeff Moden wrote:Good Lord! I've got more testing to do but I ran a simple "Tally CTE Function" test in 2017 and 2022 on my laptop. The results are REALLY disappointing....
We have seen this before. 😎 Performance concernce – SQLServerCentral Forums
I'd completely forgotten about that post, Eirikur. Thanks for the reminder. I still can't believe that such simple code is also being affected so dramatically. I even did the test with using 1 as a constant and compared that just to make sure it wasn't just a dent in ROW_NUMBER function.
--Jeff Moden
Change is inevitable... Change for the better is not.
September 14, 2023 at 1:20 pm
Hey Jeff,
I was talking to Bob Ward about your tests. He was interested in an introduction to talk about them. Any interest? If so, which email should I use?
"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
September 14, 2023 at 3:48 pm
Eirikur Eiriksson wrote:Jeff Moden wrote:Good Lord! I've got more testing to do but I ran a simple "Tally CTE Function" test in 2017 and 2022 on my laptop. The results are REALLY disappointing....
We have seen this before. 😎 Performance concernce – SQLServerCentral Forums
I'd completely forgotten about that post, Eirikur. Thanks for the reminder. I still can't believe that such simple code is also being affected so dramatically. I even did the test with using 1 as a constant and compared that just to make sure it wasn't just a dent in ROW_NUMBER function.
I'll rig up (when I have the time) the same test as I did in 2018, including the 2017 to 2022 and report the findings.
😎
September 14, 2023 at 3:56 pm
Grant Fritchey wrote:Hey Jeff,
I was talking to Bob Ward about your tests. He was interested in an introduction to talk about them. Any interest? If so, which email should I use?
I would like to be a fly on the wall in that conversation if possible Grant! 😎
Me too.
Michael L John
If you assassinate a DBA, would you pull a trigger?
To properly post on a forum:
http://www.sqlservercentral.com/articles/61537/
September 15, 2023 at 10:14 am
Eirikur Eiriksson wrote:Grant Fritchey wrote:Hey Jeff,
I was talking to Bob Ward about your tests. He was interested in an introduction to talk about them. Any interest? If so, which email should I use?
I would like to be a fly on the wall in that conversation if possible Grant! 😎
Me too.
Hell, me too. Ha!
"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
September 15, 2023 at 8:57 pm
Hey Jeff,
I was talking to Bob Ward about your tests. He was interested in an introduction to talk about them. Any interest? If so, which email should I use?
Yes. Do you know my email address that ends with ".net"? If not, I'll p.m it to you. Thanks, Grant.
--Jeff Moden
Change is inevitable... Change for the better is not.
September 15, 2023 at 9:06 pm
Thanks for the post about SQL 2022, Jeff. I've reached out to my team and mentioned that we need to fix all the problems we're currently having with SQL 2017 and 2019 ASAP before we push anyone on to 2022 just to make sure none of our code adds to the problem.
Sigh.
I wouldn't rush people, Brandie. Bad things happen when people rush. Like I say in one of my mantras, "If you want something real bad, that's the way you'll usually get it".
Eirikur and Ed (Wagner) have both confirmed the same issues in 2019 so nothing to do there. I've confirmed that there are no settings changes that make a difference compared to 2017. I'd install the 2017 stuff on a 2022 box and do some reasonable regression checking to see what is going to slow down and by how much before you start trying to fix the world of code on your side. While shooting a gun into the air is "fun", it's much more productive to have a target. 😀
--Jeff Moden
Change is inevitable... Change for the better is not.
September 15, 2023 at 9:10 pm
Jeff Moden wrote:Eirikur Eiriksson wrote:Jeff Moden wrote:Good Lord! I've got more testing to do but I ran a simple "Tally CTE Function" test in 2017 and 2022 on my laptop. The results are REALLY disappointing....
We have seen this before. 😎 Performance concernce – SQLServerCentral Forums
I'd completely forgotten about that post, Eirikur. Thanks for the reminder. I still can't believe that such simple code is also being affected so dramatically. I even did the test with using 1 as a constant and compared that just to make sure it wasn't just a dent in ROW_NUMBER function.
I'll rig up (when I have the time) the same test as I did in 2018, including the 2017 to 2022 and report the findings. 😎
Awesome, Eirikur! Ed found little change between 2019 and 2022 but having another party confirm that with different tests makes for a more complete buffet of proof by data.
--Jeff Moden
Change is inevitable... Change for the better is not.
September 16, 2023 at 4:38 pm
Grant Fritchey wrote:Hey Jeff,
I was talking to Bob Ward about your tests. He was interested in an introduction to talk about them. Any interest? If so, which email should I use?
Yes. Do you know my email address that ends with ".net"? If not, I'll p.m it to you. Thanks, Grant.
Yeah, PM me bud. I don't have that email. I'll intro & drop out (despite wanting to be a fly on the wall).
"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
September 16, 2023 at 9:13 pm
Jeff Moden wrote:Grant Fritchey wrote:Hey Jeff,
I was talking to Bob Ward about your tests. He was interested in an introduction to talk about them. Any interest? If so, which email should I use?
Yes. Do you know my email address that ends with ".net"? If not, I'll p.m it to you. Thanks, Grant.
Yeah, PM me bud. I don't have that email. I'll intro & drop out (despite wanting to be a fly on the wall).
PM sent to your SSC user message area. Thanks, Grant.
--Jeff Moden
Change is inevitable... Change for the better is not.
September 18, 2023 at 7:24 pm
On a personal note, I just don't understand companies that post a webinar and, to register, they REQUIRE you have to provide your phone number and, frequently, a lot more information that I prefer to not expose (and so I won't). I can understand needing an email to contact me with changes and cancellations but the default should be to not share such information and not automatically impose cookies rather than being able to "withdraw at any time". That includes the offerings that have been coming from our benefactor, RedGate.
If the event impresses me, make sure the YOU, the company, have exposed YOUR contact information so that I can contact YOU!
This is to let sales people get in contact. I frequently register for webinars with the denver phone 303-333-3333
September 19, 2023 at 8:02 pm
Jeff Moden wrote:On a personal note, I just don't understand companies that post a webinar and, to register, they REQUIRE you have to provide your phone number and, frequently, a lot more information that I prefer to not expose (and so I won't). I can understand needing an email to contact me with changes and cancellations but the default should be to not share such information and not automatically impose cookies rather than being able to "withdraw at any time". That includes the offerings that have been coming from our benefactor, RedGate.
If the event impresses me, make sure the YOU, the company, have exposed YOUR contact information so that I can contact YOU!
This is to let sales people get in contact. I frequently register for webinars with the denver phone 303-333-3333
The trouble with that is that it requires me to lie, which I won't do and so I don't "go".
--Jeff Moden
Change is inevitable... Change for the better is not.
September 20, 2023 at 4:06 pm
Jeff Moden wrote:On a personal note, I just don't understand companies that post a webinar and, to register, they REQUIRE you have to provide your phone number and, frequently, a lot more information that I prefer to not expose (and so I won't). I can understand needing an email to contact me with changes and cancellations but the default should be to not share such information and not automatically impose cookies rather than being able to "withdraw at any time". That includes the offerings that have been coming from our benefactor, RedGate.
If the event impresses me, make sure the YOU, the company, have exposed YOUR contact information so that I can contact YOU!
This is to let sales people get in contact. I frequently register for webinars with the denver phone 303-333-3333
I usually register for seminars and such using the pseudonym "Steve Jones". Which may explain some of the strange packages you get Steve.
Michael L John
If you assassinate a DBA, would you pull a trigger?
To properly post on a forum:
http://www.sqlservercentral.com/articles/61537/
Viewing 15 posts - 66,496 through 66,510 (of 66,709 total)
You must be logged in to reply to this topic. Login to reply