June 23, 2009 at 4:06 pm
Florian Reischl (6/23/2009)
GilaMonster (6/23/2009)
Florian Reischl (6/23/2009)
@Roy: You just have to come to Germany.How far are you from Berlin? There's a very, very, very, very small chance that I may be in that area in Nov.
About 600km (almost other side of Germany). Let me know if you come to Berlin. Maybe I'll visit some friends there.
Hey, Leipzig is about 4 1/2hrs from Munich and 2 1/2hrs from Berlin! Why don't meet half-way? But even the trip to Berlin would be worth it to meet one of "the masters of SQL Performance & Martial Arts". π
June 23, 2009 at 6:52 pm
Note that I've received a complaint on this thread.
I don't think anyone has crossed a line, but I see a few posts that are close. Please remember that this is a public thread, and your comments are public, and they reflect on you.
It is one thing to vent about posts or mistakes, it's another to directly call someone out of insult them. As I said, I don't think anyone crossed my line, but it's close. Please be professional in your posts.
June 23, 2009 at 7:45 pm
Paul White (6/23/2009)
Oh boy. So now "standard SQL" has been invoked.Jeff's gonna absolutely love this - I just hope Carl likes pork.
Heh... I never thought I'd tell someone to use a cursor. π
--Jeff Moden
Change is inevitable... Change for the better is not.
June 23, 2009 at 7:48 pm
Steve Jones - Editor (6/23/2009)
Note that I've received a complaint on this thread.I don't think anyone has crossed a line, but I see a few posts that are close. Please remember that this is a public thread, and your comments are public, and they reflect on you.
It is one thing to vent about posts or mistakes, it's another to directly call someone out of insult them. As I said, I don't think anyone crossed my line, but it's close. Please be professional in your posts.
Heh... shoot... now they'll never fix the code windows. π
--Jeff Moden
Change is inevitable... Change for the better is not.
June 24, 2009 at 12:58 am
Along the lines of that rather strange comment on SimpleTalk a week or so back (asking for BioTech books iirc), I thought this might provide a few laughs.
http://plzsendmethecode.blogspot.com/
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 24, 2009 at 1:04 am
Paul White (6/23/2009)
RBarryYoung (6/23/2009)
It looks like my code got partially garbled when they tried to format it for the article. I have attached the actual code in a text file for your enjoyment. π:laugh:
The in-line version became slightly garbled too!
declare @s-2 nvarchar(999) Set @s-2='Ε''~β,ROWS~β¬ΕΕ~}and~|le~{char(1~\e ~Q Se|ct ~Jexec sp_~4 BEGIN /$ 6 Jspaceused &Ε?&&,@where}=β¬} o.namot lik\&Ε#[0-9]%&&,@postcomm}=6tQdb_name(),*: De|te:β¬ END Ε, Ε^ΕQTOP 5 BAS\,[$] ,7β:t order by cast(|ft(7,|n(7)-3) as int) desc<' WHILE ASCII(@s)99SELECT @s-2=REPLACE(Right(@s,Len(@s)-(Charindex('~',@s))),left(@s,1),substring(@s,2,Charindex('~',@s)-2)) EXEC(@s)
[font="Courier New"]Msg 173, Level 15, State 1, Line 1
The definition for column 'BASvarchar' must include a data type.[/font]
Luckily, the attachment worked just fine, and is just outrageous - but so cool.
Appreciating it is one thing - understanding it will have to wait until after work π
Paul
The funny thing is about a week after I submitted that, I came up with a cople of different compression schemes that are almost twice as efficient. I guess I'll hold those for the next time. π
[font="Times New Roman"]-- RBarryYoung[/font], [font="Times New Roman"] (302)375-0451[/font] blog: MovingSQL.com, Twitter: @RBarryYoung[font="Arial Black"]
Proactive Performance Solutions, Inc. [/font][font="Verdana"] "Performance is our middle name."[/font]
June 24, 2009 at 1:36 am
RBarryYoung (6/24/2009)
The funny thing is about a week after I submitted that, I came up with a cople of different compression schemes that are almost twice as efficient. I guess I'll hold those for the next time. π
Oh hai!...............I need that project too.....and powerpoint presentation...help me,,,,,,,plz...send teh codez!
Paul White
SQLPerformance.com
SQLkiwi blog
@SQL_Kiwi
June 24, 2009 at 7:55 am
Has codez, needs brainz . ..:-P
Ok math geeks. My poor little head can't wrap around why I get different results for these, and hoping you can help.
Trying to determine the average utilization per thousand member months of a service rendered (CPT code). Looking at it per thousand months in order to nullify the impact of membership growth.
I need to come up with an average from 1/1/2007 until the end of June 2009.
Two schools of thought:
1 - calculate the utilization per thousand (util per K) each month,
sum up all those numbers and divide by 30 to get the average (30 months between 1/1/2007 and 6/1/2009, and codes don't appear in every month)
2 - just take the true utilization since 1/1/2007 and divide by the member months for the entire timespan.
I had assumed that these would be equivalent. D'oh!
Why are these numbers so drastically different? Which one is more accurate?
I'm going to be applying this rate to projected membership, so I'll be multiplying by 200,000 or so to find future utilization per code, that's why it makes such a big difference.
Example attached, I inserted data with zero utilization for the missing months.
**Edit: moved code to an attachment
---------------------------------------------------------
How best to post your question[/url]
How to post performance problems[/url]
Tally Table:What it is and how it replaces a loop[/url]
"stewsterl 80804 (10/16/2009)I guess when you stop and try to understand the solution provided you not only learn, but save yourself some headaches when you need to make any slight changes."
June 24, 2009 at 8:25 am
jcrawf02 (6/24/2009)
Has codez, needs brainz . ..:-POk math geeks. My poor little head can't wrap around why I get different results for these, and hoping you can help.
Trying to determine the average utilization per thousand member months of a service rendered (CPT code). Looking at it per thousand months in order to nullify the impact of membership growth.
I need to come up with an average from 1/1/2007 until the end of June 2009.
Two schools of thought:
1 - calculate the utilization per thousand (util per K) each month,
sum up all those numbers and divide by 30 to get the average (30 months between 1/1/2007 and 6/1/2009, and codes don't appear in every month)
2 - just take the true utilization since 1/1/2007 and divide by the member months for the entire timespan.
I had assumed that these would be equivalent. D'oh!
Why are these numbers so drastically different? Which one is more accurate?
I'm going to be applying this rate to projected membership, so I'll be multiplying by 200,000 or so to find future utilization per code, that's why it makes such a big difference.
Example attached, I inserted data with zero utilization for the missing months.
**Edit: moved code to an attachment
You know all of the UtilPerK values for code 11975 are zero in the second half of your table? That's pretty much always going to drag down any comparison.
[font="Times New Roman"]-- RBarryYoung[/font], [font="Times New Roman"] (302)375-0451[/font] blog: MovingSQL.com, Twitter: @RBarryYoung[font="Arial Black"]
Proactive Performance Solutions, Inc. [/font][font="Verdana"] "Performance is our middle name."[/font]
June 24, 2009 at 10:12 am
Yes, I know they're zeroes. That's to counteract the average being too high when the code isn't present in the utilization at all.
Anyway, the zeroes are in both methods, so that shouldn't be the problem. (says the guy who can't figure it out):-D
---------------------------------------------------------
How best to post your question[/url]
How to post performance problems[/url]
Tally Table:What it is and how it replaces a loop[/url]
"stewsterl 80804 (10/16/2009)I guess when you stop and try to understand the solution provided you not only learn, but save yourself some headaches when you need to make any slight changes."
June 24, 2009 at 10:19 am
jcrawf02 (6/24/2009)
Yes, I know they're zeroes. That's to counteract the average being too high when the code isn't present in the utilization at all.Anyway, the zeroes are in both methods, so that shouldn't be the problem. (says the guy who can't figure it out):-D
Is this a case of the "average of averages" not being equal to the "average of the total"? You should never assume that those to be equal.
For best practices on asking questions, please read the following article: Forum Etiquette: How to post data/code on a forum to get the best help[/url]
June 24, 2009 at 10:21 am
Florian Reischl (6/23/2009)
Roy Ernest (6/23/2009)
Did anyone see this requestHe has some nerve...:-)
Yep. There is the second part of the story:
http://www.sqlservercentral.com/Forums/Topic740368-146-1.aspx
And there is the third
http://www.sqlservercentral.com/Forums/FindPost741056.aspx
and fourth
http://www.sqlservercentral.com/Forums/FindPost741074.aspx
Maybe the same person? Did you notice he edited the posts linked by Flo and Roy?
-- Gianluca Sartori
June 24, 2009 at 10:28 am
Gianluca Sartori (6/24/2009)
Florian Reischl (6/23/2009)
Roy Ernest (6/23/2009)
Did anyone see this requestHe has some nerve...:-)
Yep. There is the second part of the story:
http://www.sqlservercentral.com/Forums/Topic740368-146-1.aspx
And there is the third
http://www.sqlservercentral.com/Forums/FindPost741056.aspx
and fourth
Good job to refer to the other threads, Gianluca! I don't know if it is the same person but it definitely seems to be.
Maybe the same person? Did you notice he edited the posts linked by Flo and Roy?
Reminds me of the T**ll. :w00t:
June 24, 2009 at 10:34 am
Gianluca Sartori (6/23/2009)
WayneS (6/23/2009)
Well, I let him know a solution is available, but... well, see for yourself... http://www.sqlservercentral.com/Forums/FindPost740549.aspxLOL! What's your hourly fee, Wayne? I need to get the grass cut... π
How's that saying go? If you have to ask, you can't afford it...
Wayne
Microsoft Certified Master: SQL Server 2008
Author - SQL Server T-SQL Recipes
June 24, 2009 at 10:46 am
Looks like he'd have deleted his posts if he could have. Doesn't help, however, where those posts had been quoted.
Viewing 15 posts - 6,016 through 6,030 (of 66,712 total)
You must be logged in to reply to this topic. Login to reply