June 2, 2016 at 2:18 pm
jasona.work (6/2/2016)
I read this and suddenly felt millions of DBAs cry out in horror...(Yes, it's the Daily WTF, scarily it might even be (mostly) true)
ROFL. That is awesome. I wouldn't be surprised if is at least based on some truths. Like an old school developer with zero clue about relational data did something like that. There is little chance that a system like ever actually made it to production. And if it did they certainly wouldn't be ok with "weeks leading up to his visit to the client site". Still it is hilarious.
_______________________________________________________________
Need help? Help us help you.
Read the article at http://www.sqlservercentral.com/articles/Best+Practices/61537/ for best practices on asking questions.
Need to split a string? Try Jeff Modens splitter http://www.sqlservercentral.com/articles/Tally+Table/72993/.
Cross Tabs and Pivots, Part 1 – Converting Rows to Columns - http://www.sqlservercentral.com/articles/T-SQL/63681/
Cross Tabs and Pivots, Part 2 - Dynamic Cross Tabs - http://www.sqlservercentral.com/articles/Crosstab/65048/
Understanding and Using APPLY (Part 1) - http://www.sqlservercentral.com/articles/APPLY/69953/
Understanding and Using APPLY (Part 2) - http://www.sqlservercentral.com/articles/APPLY/69954/
June 2, 2016 at 2:19 pm
jasona.work (6/2/2016)
I read this and suddenly felt millions of DBAs cry out in horror...(Yes, it's the Daily WTF, scarily it might even be (mostly) true)
Great story. A true horror. And I'm calling shenanigans.
----------------------------------------------------The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood... Theodore RooseveltThe Scary DBAAuthor of: SQL Server 2017 Query Performance Tuning, 5th Edition and SQL Server Execution Plans, 3rd EditionProduct Evangelist for Red Gate Software
June 2, 2016 at 4:26 pm
Grant Fritchey (6/2/2016)
jasona.work (6/2/2016)
I read this and suddenly felt millions of DBAs cry out in horror...(Yes, it's the Daily WTF, scarily it might even be (mostly) true)
Great story. A true horror. And I'm calling shenanigans.
I'd like to do the same, but I know of an actual DB implementation that does something fairly similar.
One record per customer, in a 2 column table (id and path to file), all customer's contracts and other details in an XML blob stored in the file system. App pulls the entire blob up, shreds and searches it in memory.
Performance = abysmal. Data quality = what's that. Architects = use the file system because the database is too slow.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 2, 2016 at 4:37 pm
GilaMonster (6/2/2016)
Grant Fritchey (6/2/2016)
jasona.work (6/2/2016)
I read this and suddenly felt millions of DBAs cry out in horror...(Yes, it's the Daily WTF, scarily it might even be (mostly) true)
Great story. A true horror. And I'm calling shenanigans.
I'd like to do the same, but I know of an actual DB implementation that does something fairly similar.
One record per customer, in a 2 column table (id and path to file), all customer's contracts and other details in an XML blob stored in the file system. App pulls the entire blob up, shreds and searches it in memory.
Performance = abysmal. Data quality = what's that. Architects = use the file system because the database is too slow.
Wow!
And the explanation... "SQL Server sucks!"
----------------------------------------------------The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood... Theodore RooseveltThe Scary DBAAuthor of: SQL Server 2017 Query Performance Tuning, 5th Edition and SQL Server Execution Plans, 3rd EditionProduct Evangelist for Red Gate Software
June 3, 2016 at 12:20 am
Grant Fritchey (6/2/2016)
Wow!And the explanation... "SQL Server sucks!"
Oracle, not SQL Server in this case, but kinda yes.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 3, 2016 at 1:39 am
jasona.work (6/2/2016)
I read this and suddenly felt millions of DBAs cry out in horror...(Yes, it's the Daily WTF, scarily it might even be (mostly) true)
This story probably is true because there are many examples to choose from. Couple of years ago I worked for a pensions company who at the time were evaluating a new third-party product to replace their aging mainframe. Everything worked fine and dandy except it didn't scale. Tests showed that the nightly process would take over 24 hours. We visited the vendor and dug a little deeper into their design, and found that child rows (and their child rows) were held in parent tables as huge xml structures.
To paraphrase someone's tagline, just because you can do something, doesn't mean you should.
For fast, accurate and documented assistance in answering your questions, please read this article.
Understanding and using APPLY, (I) and (II) Paul White
Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden
June 3, 2016 at 1:46 am
Phil Parkin (6/2/2016)
BLOB_EATER (6/2/2016)
Luis Cazares (6/2/2016)
jasona.work (6/2/2016)
Happy National Leave the Office Early Day!!![/url]So, who's going to celebrate this day today?
😀
After leaving the office after 8pm everyday for 2 years, I've been celebrating this everyday for the last year. 😀
Somehow, 4pm seems to early to leave the office.
The looks I get for leaving at 4pm, yet they do not know that I walk in at 7am.
This is always a problem for early-starters.
I suggest that you return the favour to those who just about manage to rumble in at 9am.
That's me too. I like to tell the users that they gave and performance issues before they log in rather than them tell me.
-------------------------------Posting Data Etiquette - Jeff Moden [/url]Smart way to ask a question
There are naive questions, tedious questions, ill-phrased questions, questions put after inadequate self-criticism. But every question is a cry to understand (the world). There is no such thing as a dumb question. ― Carl Sagan
I would never join a club that would allow me as a member - Groucho Marx
June 3, 2016 at 1:47 am
GilaMonster (6/2/2016)
Grant Fritchey (6/2/2016)
jasona.work (6/2/2016)
I read this and suddenly felt millions of DBAs cry out in horror...(Yes, it's the Daily WTF, scarily it might even be (mostly) true)
Great story. A true horror. And I'm calling shenanigans.
I'd like to do the same, but I know of an actual DB implementation that does something fairly similar.
One record per customer, in a 2 column table (id and path to file), all customer's contracts and other details in an XML blob stored in the file system. App pulls the entire blob up, shreds and searches it in memory.
Performance = abysmal. Data quality = what's that. Architects = use the file system because the database is too slow.
This could easily be the pension system I mentioned ^^. The company I was with didn't buy into it for obvious reasons but other companies did. It's live, it's out there, and it's probably still being sold. I can't bear to look. It's Friday. Happy Friday.
Edit: nope, yours is Oracle. Mine's DB2.
For fast, accurate and documented assistance in answering your questions, please read this article.
Understanding and using APPLY, (I) and (II) Paul White
Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden
June 3, 2016 at 1:57 am
Anyone seen the Linux SQL preview on channel 9? I like it. Once you installed the tools something like:
(login first).... then:
> ~$ sudo apt-get install mssql-server
> ~$ sudo apt-get install mssql-tools
> ~$ sqlcmd < enter commands here>
feels just like home, also, they say you can use SSMS to connect to the instance above too. Pretty cool eh?
June 3, 2016 at 4:35 am
Lynn Pettis (6/1/2016)
Just looking at the diagrams here does anyone else think a few of them may be wrong? It could just be me, but I need to do some testing to put my mind at ease.
Which ones do you think are wrong? The LEFT & RIGHT JOINs with the key is null in the other table seem right, as do most of the others. The only one that looks weird to me is the FULL OUTER where one key or another is NULL. The OR should (IMHO) make it shaded all the way across.
June 3, 2016 at 4:39 am
Hugo Kornelis (6/1/2016)
Lynn Pettis (6/1/2016)
Just looking at the diagrams here does anyone else think a few of them may be wrong? It could just be me, but I need to do some testing to put my mind at ease.The diagrams are correct, but I do not like the use of outer join + IS NULL to replace what should be a simple NOT EXISTS plus correlated subquery.
The lower right one looks odd, I cannot imagine ever needing that in a real situation.
Actually, I use the outer join + IS NULL quite frequently usually because that is the simpler way of doing the query in my workplace.
I can't see using the FULL OUTER at all, either configuration, in a real situation. But I agree the lower right one looks odd.
June 3, 2016 at 4:41 am
Steve Jones - SSC Editor (6/1/2016)
SSMS is decoupled from SQL. You need to install that separately.
Wait. Client tools are no longer an option to install with the rest of SQL Server?
That sucks.
June 3, 2016 at 4:43 am
ChrisM@Work (6/3/2016)
GilaMonster (6/2/2016)
Grant Fritchey (6/2/2016)
jasona.work (6/2/2016)
I read this and suddenly felt millions of DBAs cry out in horror...(Yes, it's the Daily WTF, scarily it might even be (mostly) true)
Great story. A true horror. And I'm calling shenanigans.
I'd like to do the same, but I know of an actual DB implementation that does something fairly similar.
One record per customer, in a 2 column table (id and path to file), all customer's contracts and other details in an XML blob stored in the file system. App pulls the entire blob up, shreds and searches it in memory.
Performance = abysmal. Data quality = what's that. Architects = use the file system because the database is too slow.
This could easily be the pension system I mentioned ^^. The company I was with didn't buy into it for obvious reasons but other companies did. It's live, it's out there, and it's probably still being sold. I can't bear to look. It's Friday. Happy Friday.
Edit: nope, yours is Oracle. Mine's DB2.
Mine isn't a vendor app either. It's custom-built by a large local company, a friend of mine works as a developer there. The 'architects' used all sorts of buzzwords to justify the design about 5 years ago, and they still refuse to accept that it's not ideal.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 3, 2016 at 4:46 am
Brandie Tarvin (6/3/2016)
Steve Jones - SSC Editor (6/1/2016)
SSMS is decoupled from SQL. You need to install that separately.Wait. Client tools are no longer an option to install with the rest of SQL Server?
That sucks.
Correct, it's a separate install.
Personally, I love it. Fewer prod servers with full client tools, fewer laptops with full SQL Server on when they just need SSMS. It's not a lot more work to install.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 3, 2016 at 4:57 am
Brandie Tarvin (6/3/2016)
Lynn Pettis (6/1/2016)
Just looking at the diagrams here does anyone else think a few of them may be wrong? It could just be me, but I need to do some testing to put my mind at ease.Which ones do you think are wrong? The LEFT & RIGHT JOINs with the key is null in the other table seem right, as do most of the others. The only one that looks weird to me is the FULL OUTER where one key or another is NULL. The OR should (IMHO) make it shaded all the way across.
No, it's correct. The middle area is where both keys have a non-null value. (Assuming it's a non-nullable column)
Viewing 15 posts - 54,421 through 54,435 (of 66,000 total)
You must be logged in to reply to this topic. Login to reply