February 25, 2016 at 6:50 am
GilaMonster (2/24/2016)
I disagree. Jupiter is descending.
That is an awesome link. Thanks, Gail.
On another note, this might be one of those days when I should step away from my computer. I get back from a week & 1/2 of vacation and the first thing I do is mix up three projects and try to do coding on one that should be done on another... Sigh. Jupiter is a nice distraction so I can reorder my thoughts and actually remember which project is doing what thing.
Good thing I made notes before I left. Bad thing I used shorthand, thinking I'd remember what the heck I meant.
February 25, 2016 at 7:30 am
jasona.work (2/24/2016)
SQLRNNR (2/24/2016)
GilaMonster (2/24/2016)
jasona.work (2/24/2016)
http://www.sqlservercentral.com/Forums/Topic1763775-3412-1.aspx/Me leans back, grabs bucket of popcorn
Sure sounds like Gail answered the original question, yet for some reason the OP decided to focus on the comments about performance...
INSERT INTO UsersToBeIgnored...
Though, I might have to partition that table, it's running the current file out of space. π
Then would you have multiple files in the partition? Would you span them across a gpt formatted volume? Should the volume be offset at 17.5k or something else?
What are the performance implications?
So what you are saying is the end result is that the partitioned table sucks for performance? π
What if we have 8 files per partition per database? Would that make performance better?
/Me hands Gail a wet fish, steps back out of the splatter zone...
:hehe:
Thanks... it's been too long since I've had such a good laugh.
Wayne
Microsoft Certified Master: SQL Server 2008
Author - SQL Server T-SQL Recipes
February 25, 2016 at 7:39 am
Is it too early for these answers? http://www.sqlservercentral.com/Forums/Topic1764225-3411-1.aspx
February 25, 2016 at 7:43 am
Luis Cazares (2/25/2016)
Is it too early for these answers? http://www.sqlservercentral.com/Forums/Topic1764225-3411-1.aspx
Not at all.
The absence of evidence is not evidence of absence
- Martin Rees
The absence of consumable DDL, sample data and desired results is, however, evidence of the absence of my response
- Phil Parkin
February 25, 2016 at 9:29 am
Phil Parkin (2/25/2016)
Luis Cazares (2/25/2016)
Is it too early for these answers? http://www.sqlservercentral.com/Forums/Topic1764225-3411-1.aspxNot at all.
At least you didn't go with LMGTFY, or JFGI...
Thomas Rushton
blog: https://thelonedba.wordpress.com
February 25, 2016 at 9:31 am
ThomasRushton (2/25/2016)
Phil Parkin (2/25/2016)
Luis Cazares (2/25/2016)
Is it too early for these answers? http://www.sqlservercentral.com/Forums/Topic1764225-3411-1.aspxNot at all.
At least you didn't go with LMGTFY, or JFGI...
That would have implied additional work.
February 25, 2016 at 9:39 am
I am so glad I don't have any experience, yet, with replication so that I have no idea what SS is trying to do or why he isn't getting what he expects.
February 25, 2016 at 9:45 am
Lynn Pettis (2/25/2016)
I am so glad I don't have any experience, yet, with replication so that I have no idea what SS is trying to do or why he isn't getting what he expects.
Wow. He really didn't include any details on what he's trying to accomplish.
I've created stuff to deliberately fail before, but usually after I've done every other bit of troubleshooting I can think of.
February 25, 2016 at 1:27 pm
Man, my music tastes go all over the place...
Listening to my playlist on my phone today, just went from Dragonforce, now I'm listening to an Alex Band tune w/Santana, and the whole thing started off with Motley Crue...
Who knows, maybe Weird Al next...
February 25, 2016 at 1:41 pm
Grant Fritchey (2/24/2016)
Jeff Moden (2/24/2016)
SQLRNNR (2/24/2016)
GilaMonster (2/24/2016)
SQLRNNR (2/24/2016)
GilaMonster (2/24/2016)
jasona.work (2/24/2016)
http://www.sqlservercentral.com/Forums/Topic1763775-3412-1.aspx/Me leans back, grabs bucket of popcorn
Sure sounds like Gail answered the original question, yet for some reason the OP decided to focus on the comments about performance...
INSERT INTO UsersToBeIgnored...
Though, I might have to partition that table, it's running the current file out of space. π
Then would you have multiple files in the partition? Would you span them across a gpt formatted volume? Should the volume be offset at 17.5k or something else?
What are the performance implications?
So what you are saying is the end result is that the partitioned table sucks for performance? π
What if we have 8 files per partition per database? Would that make performance better?
Depends on what time sunset is and which planets are ascending.
Jupiter Ascending
Earth Ascending inversely
Pork Chops Away!:-D
I get really tired of these moroffs blasting the likes of Gail when they don't actually have a clue. Hope he likes his own medicine. π
I swear we've had almost this exact conversation with identical jokes before.
Many times. That's because it's the same old story with some of these folks. Hopefully, my response on that thread got his attention.
--Jeff Moden
Change is inevitable... Change for the better is not.
February 25, 2016 at 1:42 pm
jasona.work (2/25/2016)
Man, my music tastes go all over the place...Listening to my playlist on my phone today, just went from Dragonforce, now I'm listening to an Alex Band tune w/Santana, and the whole thing started off with Motley Crue...
Who knows, maybe Weird Al next...
Went from "Jesus Is Risen Today" to "Layla" a little later Weird Al...
February 25, 2016 at 1:52 pm
Spent a couple hours rewriting a query that was taking 4 hours. Got it down to 8 minutes! Then the light bulb goes off and I realize my code does not do what the other code did. :hehe: I have corrected the issue with a CROSS APPLY and am now getting the 66 million expected records (17 minutes), however still I need to insert them into the final table where I just changed the clustered key to be sensible. (6 minutes and counting)
February 25, 2016 at 2:12 pm
djj (2/25/2016)
Spent a couple hours rewriting a query that was taking 4 hours. Got it down to 8 minutes! Then the light bulb goes off and I realize my code does not do what the other code did. :hehe:
+1000
Been there, done that... I hate when that happens! π
-- Itzik Ben-Gan 2001
February 25, 2016 at 2:21 pm
Alan.B (2/25/2016)
djj (2/25/2016)
Spent a couple hours rewriting a query that was taking 4 hours. Got it down to 8 minutes! Then the light bulb goes off and I realize my code does not do what the other code did. :hehe:+1000
Been there, done that... I hate when that happens! π
Yeah, me too. But when you get it right and you watch the 3 hour query go down to 2 seconds, it sure is nice. Or even better, when you get 8-second query that gets run 30,000 times a day down to 50 ms.
February 26, 2016 at 1:17 am
djj (2/25/2016)
jasona.work (2/25/2016)
Man, my music tastes go all over the place...Listening to my playlist on my phone today, just went from Dragonforce, now I'm listening to an Alex Band tune w/Santana, and the whole thing started off with Motley Crue...
Who knows, maybe Weird Al next...
Went from "Jesus Is Risen Today" to "Layla" a little later Weird Al...
I've gone from Cannibal Corpse to Pink Floyd while out running before. That can upset the rhythm a bit, particularly at the bottom of the climb to get home.
How to post a question to get the most help http://www.sqlservercentral.com/articles/Best+Practices/61537
Viewing 15 posts - 52,906 through 52,920 (of 66,712 total)
You must be logged in to reply to this topic. Login to reply