Are the posted questions getting worse?
-
According to Paul White (and what I heard early on when they first started meta-caching of Temp Tables), dropping the tables does NOT prevent caching of Temp Table meta-data.
https://sqlperformance.com/2017/05/sql-performance/sql-server-temporary-object-caching
I will admit that it's a waste of code space unless, like I said before, you're using a really big Temp Table and you just want to free up some space in TempDB because you're done with the large temp table but your proc isn't done yet. It's still going to have the meta-data cached. You could also just truncate the table to do the same thing.
--Jeff Moden
RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.Change is inevitable... Change for the better is not.
Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally) -
Got the first post on a page again... and again, it's not showing up until another post (this one) is made.
--Jeff Moden
RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.Change is inevitable... Change for the better is not.
Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally) - Jeff Moden wrote:
Got the first post on a page again... and again, it's not showing up until another post (this one) is made.
I've seen that behaviour too.
The absence of evidence is not evidence of absence.
Martin ReesYou can lead a horse to water, but a pencil must be lead.
Stan Laurel -
Shifting gears a bit, I'm hoping this year that work will be happy to pay (or at least contribute) to me getting out and going to some SQL based events this year; it's something I've been trying to get out of them for a while, and it seems that the badgering is finally bearing some fruit.
Now that things are mostly back to normal, after lockdowns, etc, I'm hoping that there will be some choice this year, however, does anyone have any particular ones that they would recommend, and know (or hope) are going ahead this year for someone who's UK based? I'm, personally, happy to travel more or less anywhere within the country, but I'm based near London if they helps or changes any of your recommendations.
If you do know of any big events over in Europe give them a shout too, but I suspect I might find it harder for work to pay outright (but if it's a good event, and I can get them to contribute, then it might still be on the cards for me).
Thanks for any ideas you might have.
Thom~
Excuse my typos and sometimes awful grammar. My fingers work faster than my brain does.
Larnu.uk -
Talking about ChatGPT, has anyone seen this one yet? https://www.cnn.com/2023/02/08/tech/google-ai-bard-demo-error/index.html
<headdesk>
- Thom A wrote:
Shifting gears a bit, I'm hoping this year that work will be happy to pay (or at least contribute) to me getting out and going to some SQL based events this year; it's something I've been trying to get out of them for a while, and it seems that the badgering is finally bearing some fruit.
Now that things are mostly back to normal, after lockdowns, etc, I'm hoping that there will be some choice this year, however, does anyone have any particular ones that they would recommend, and know (or hope) are going ahead this year for someone who's UK based? I'm, personally, happy to travel more or less anywhere within the country, but I'm based near London if they helps or changes any of your recommendations.
If you do know of any big events over in Europe give them a shout too, but I suspect I might find it harder for work to pay outright (but if it's a good event, and I can get them to contribute, then it might still be on the cards for me).
Thanks for any ideas you might have.
SQLBits, but call for speakers closed months ago. That's the big one in the UK. Next is Relay. The call for speakers probably opens late spring. After that, Data Ceilie (sp?) in Ireland and then various Data & SQL Saturday events.
I can list some in europe too if you want.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore RooseveltAuthor of:
SQL Server Execution Plans
SQL Server Query Performance Tuning - Grant Fritchey wrote:
SQLBits, but call for speakers closed months ago. That's the big one in the UK. Next is Relay. The call for speakers probably opens late spring. After that, Data Ceilie (sp?) in Ireland and then various Data & SQL Saturday events.
I can list some in europe too if you want.
Thanks Grant. I'm not looking at doing any speaking, just attending. I'll have a look at these and see how they fit in with the Calendar. 🙂
Thom~
Excuse my typos and sometimes awful grammar. My fingers work faster than my brain does.
Larnu.uk - Thom A wrote:
Shifting gears a bit, I'm hoping this year that work will be happy to pay (or at least contribute) to me getting out and going to some SQL based events this year; it's something I've been trying to get out of them for a while, and it seems that the badgering is finally bearing some fruit.
Now that things are mostly back to normal, after lockdowns, etc, I'm hoping that there will be some choice this year, however, does anyone have any particular ones that they would recommend, and know (or hope) are going ahead this year for someone who's UK based? I'm, personally, happy to travel more or less anywhere within the country, but I'm based near London if they helps or changes any of your recommendations.
If you do know of any big events over in Europe give them a shout too, but I suspect I might find it harder for work to pay outright (but if it's a good event, and I can get them to contribute, then it might still be on the cards for me).
Thanks for any ideas you might have.
I have the same plans Thom. I went to SQL Saturday in Richmond in October, and am going to go to SQL Saturday Atlanta in a few weeks. It was a tough call between Atlanta and Austin, they are the same day, but when I saw Grant was speaking at Austin, I picked Atlanta. 🤣🤣
SQL Bits was on the agenda, but things came up.
It appears that some of the SQL user groups and SQL Saturdays are starting to get going, as well as the various conferences. My company is split on paying for these things. There's a new VP and he's been burned before, so he's hesitant to blindly approve these things.
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/ - Jeff Moden wrote:
According to Paul White (and what I heard early on when they first started meta-caching of Temp Tables), dropping the tables does NOT prevent caching of Temp Table meta-data.
https://sqlperformance.com/2017/05/sql-performance/sql-server-temporary-object-caching
I will admit that it's a waste of code space unless, like I said before, you're using a really big Temp Table and you just want to free up some space in TempDB because you're done with the large temp table but your proc isn't done yet. It's still going to have the meta-data cached. You could also just truncate the table to do the same thing.
All of this tempdb/temptables discussion is spot on to the things I am dealing with daily. I spent half a day trying to remember where I saw some of these things that I started preaching to the developers about. SSC pulls thru again!
There is a mission-critical system here that has a boatload of SSIS packages running on it. The databases are not really data warehouses, and not really OLTP systems. The data comes out of the mainframe systems, so it starts out a mess and ends up in a slightly less of a mess.
This is an Azure VM, they opened a ticket with MS a YEAR AGO. They did the classic MS support, run PSSDIAG. They never ran it successfully. MS's only recommendation in a year was to increase the size of the VM.
Guess who whipped this baby into shape in about 2 days? There have been no package failures for 2 weeks, when there were 3-10 occurring daily. The best part is that the powers that be noticed.
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/ - Brandie Tarvin wrote:
Talking about ChatGPT, has anyone seen this one yet? https://www.cnn.com/2023/02/08/tech/google-ai-bard-demo-error/index.html
<headdesk>
Yes... and they're making an big deal about it. ChatGPT is constantly providing incorrect answers about code and people and doing so in such a manner that someone that doesn't know better will believe the damned thing.
Again, people have to remember that the "A" in "AI" stands for "Artificial"... Not "Actual" nor "Accurate".
Here's just one example "conversation" I had with it... the first answer is probably why it was able to "pass a Google interview for a $183K Level 3" programming position. Most actual people don't know the correct answer either even though the much better methods have been out for a very long time. Notice also that this is confirmation that it DOES know what a "Recursive CTE" is... that'll be very important for the followup questions I asked it.
See anything wrong in the following?
How about this one?
This one actually pissed me off. There is no way in hell I'd do it in such a fashion. I should sue for defamatory information. 😀 My point is that it's now in the "automatic bullshit grinder" mode... Ironically, that's the same thing that a lot of actual people incorrectly resort to.
--Jeff Moden
RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.Change is inevitable... Change for the better is not.
Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally) - Michael L John wrote:
SSC pulls thru again!
{snip}
Guess who whipped this baby into shape in about 2 days?
Someone who's an obvious and valued resident of the SSC Sanitorium in the DBA Disneyland section of the complex. 😀 😀 😀
Ya just gotta love us old and crazy folks. 😀
--Jeff Moden
RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.Change is inevitable... Change for the better is not.
Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally) - Jeff Moden wrote:Brandie Tarvin wrote:
Talking about ChatGPT, has anyone seen this one yet? https://www.cnn.com/2023/02/08/tech/google-ai-bard-demo-error/index.html
<headdesk>
Yes... and they're making an big deal about it. ChatGPT is constantly providing incorrect answers about code and people and doing so in such a manner that someone that doesn't know better will believe the damned thing.
You're missing the entire point of this.
I had a discussion last night with one of the behavioral psychologists from my old job, Brilliant person. He's only 28 years old and thinks on an entirely different level. His contention is that AI will never be able to replace human learning because you cannot "program random". It's always going to be an algorithm. He expects it to die after people realize that it's wrong more often than not.
My come back is that ChatGPT, etc. will become very pervasive not because it produces CORRECT output, but because the output will eventually be good enough, and the companies can make money on it. If they market it the right way, the masses will buy into it. Right or wrong, if you sell something properly, people will buy it.
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/ -
Nope... not missing those points at all. Those are the points that I'm actually afraid of because the folks that need to use it are going to believe its "correct" instead of "good enough" or "flat out wrong".
Heh... Where's "Magnus, the Robot Killer" when you need him? 😀
--Jeff Moden
RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.Change is inevitable... Change for the better is not.
Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally) - Jeff Moden wrote:
According to Paul White (and what I heard early on when they first started meta-caching of Temp Tables), dropping the tables does NOT prevent caching of Temp Table meta-data.
https://sqlperformance.com/2017/05/sql-performance/sql-server-temporary-object-caching
I will admit that it's a waste of code space unless, like I said before, you're using a really big Temp Table and you just want to free up some space in TempDB because you're done with the large temp table but your proc isn't done yet. It's still going to have the meta-data cached. You could also just truncate the table to do the same thing.
Sorry that is correct, dropping a temp table doesn't invalidate caching. I was remembering this part of the post I linked to:
Keep in mind that even with these fixes, it's still possible to hit metadata contention in tempdb. We are continuing to work on improving tempdb performance and metadata contention, but in the meantime, there are some best practices you can employ in your code that might help avoid the contention:
Do not explicitly drop temp tables at the end of a stored procedure, they will get cleaned up when the session that created them ends.
Do not alter temp tables after they have been created.
Do not truncate temp tables
Move index creation statements on temp tables to the new inline index creation syntax that was introduced in SQL Server 2014.
Jack Corbett
Consultant - Straight Path Solutions
Check out these links on how to get faster and more accurate answers:
Forum Etiquette: How to post data/code on a forum to get the best help
Need an Answer? Actually, No ... You Need a Question - Thom A wrote:
Thanks Grant. I'm not looking at doing any speaking, just attending. I'll have a look at these and see how they fit in with the Calendar. 🙂
Misunderstood. Apologies. So, Bits is in a few weeks, taking place just outside Cardiff. Saturday is free. Other days, other costs.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore RooseveltAuthor of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
Viewing 15 posts - 66,196 through 66,210 (of 66,738 total)
You must be logged in to reply to this topic. Login to reply