September 21, 2009 at 9:23 pm
Comments posted to this topic are about the item The Dynamic Tally or Numbers Table
September 22, 2009 at 1:37 am
Excellent article, Lynn!
I'll make sure to include your oustanding UDF in my dev/production environment. I already have one that does the same thing, in a while loop :blush:.
Thanks for this content, it will be very useful!
-- Gianluca Sartori
September 22, 2009 at 3:19 am
Top job Lynn.
It's this sort of article that really keeps me on my toes with respect to the SQL I write - there's almost always a better way, and this article has shown me a better way.
Really good explanation too. Thanks! 😀
Atlantis Interactive - SQL Server Tools
My blog[/url]
Why I wrote a sql query analyzer clone
September 22, 2009 at 6:16 am
I quickly read the article and ran no tests myself, but if all this code is only used to quickly fill a table with numbers then I am perplexed about the complexity as well as the amount of code. We all have seen so many 5 line SQL code solutions that do the same, be it only for tally table generation.
If the code is to dynamically generate a list of numbers in a desired range for immediate use in another statement I still find the code rather complex and would like to see comparisons with dozens of other tally solutions that can be found on this site. We all know by now that recursion is not the tool to use for this and to compare it with that doesn’t say much to me.
Don't take this the wrong way, but can you clarify to me what the exact intended use is and if you did compare it with similar solutions that are not rooted in recursion? I can see the range and order aspect as something novel compared to pure tally functions, but a function that simply reads a tally table could do the same in a single statment by using some basic math.
What required the many staged approach you took?
September 22, 2009 at 6:39 am
I agree. Definitely an interesting article but I would have liked to see tests against an already created and indexed tally table and the reasons for going dynamic rather than static since it is quite easy to sort and scale results to any interval/order from a static table.
I'll have a look myself of course but for now cannot thing of a reason to use a function over a static tally table. Maybe I'm missing something though.
[font="Arial"]_______________________________________________________
Change is inevitable... Except from a vending machine.[/font]
September 22, 2009 at 6:53 am
Here is an exampe of generating a range of numbers form an existing tally table or function with numbers starting at 1.
-- Get a range of numbers based on a tally table, the range can be 20bits wide
-- (1048576 positions) in this example.
declare @min-2 bigint, @max-2 bigint;
select @min-2 = 1000000000, @max-2 = 1000000200;
select top (@max - @min-2 + 1) (@min - 1) + N from dbo.tally20b;
Even with a 12 bit tally table you can generate ranges of 4096 numbers at any position in the bigint range. Creating intervals is no more then adding a multiplyer (see below):
-- Get a range of numbers based on a tally table, the range can be 20bits wide
-- (1048576 positions) in this example. An interval of 10 is used.
declare @min-2 bigint, @max-2 bigint;
select @min-2 = 1000000000, @max-2 = 1000000200;
select top (@max - @min-2 + 1) @min-2 + (10 * (N - 1)) from dbo.tally20b;
September 22, 2009 at 7:26 am
Rob Fisk (9/22/2009)
I agree. Definitely an interesting article but I would have liked to see tests against an already created and indexed tally table and the reasons for going dynamic rather than static since it is quite easy to sort and scale results to any interval/order from a static table.I'll have a look myself of course but for now cannot thing of a reason to use a function over a static tally table. Maybe I'm missing something though.
I can't tell you the number of times a tally or numbers table approach has been suggested to someone on the forums here and the response was something to the affect of "I'm not allowed to change the schema" or "I can't get a static table like that approved." Seems like a quick way to dynamically generate a tally table of just the numbers you need would be very helpful in those circumstances...
Good thorough description of the code Lynn, and a good read.
Thanks,
-Luke.
September 22, 2009 at 7:34 am
Indeed, a very well written, descriptive and, as I said, interesting article. It's already been bookmarked so I can look at the interaction of the various elements in the code later since it seems they could help me with other things I have been looking at.
I had completely forgotten the scenario of being able to create functions but not tables. Especially if those tables have no impact on the rest of the schema. I must have been on some happy juice and forgotten what a crazy world wee live in.
[font="Arial"]_______________________________________________________
Change is inevitable... Except from a vending machine.[/font]
September 22, 2009 at 8:28 am
I am a bit perplexed by the article. I buiul a tally table in each of my databases a couple of years ago. Static table, 100K rows, PK. When I initially built it I really did not care if it took 150ms or 1.3 seconds, because it was a one-time thing. A tally table is so useful, it should be part of the SQL installation and be included when a new database is created.
An article on how to use a tally table would have been a lot better than how to create one. Likle it was said, there are a lot of articles that provide a 5-line code solution on how to create it all over the place. No reason to be creating one on the fly every time you need it.
September 22, 2009 at 8:33 am
September 22, 2009 at 8:36 am
I like the article but just a minor point, the code doesnt work, maybe it is the way it came across when formatted.
You are missing a space, in your set statistics statements
in the code blocks like ioon instead of "io on" for example
--------------------------------------------------------------------------------------
[highlight]Recommended Articles on How to help us help you and[/highlight]
[highlight]solve commonly asked questions[/highlight]
Forum Etiquette: How to post data/code on a forum to get the best help by Jeff Moden[/url]
Managing Transaction Logs by Gail Shaw[/url]
How to post Performance problems by Gail Shaw[/url]
Help, my database is corrupt. Now what? by Gail Shaw[/url]
September 22, 2009 at 8:39 am
Silverfox (9/22/2009)
I like the article but just a minor point, the code doesnt work, maybe it is the way it came across when formatted.You are missing a space, in your set statistics statements
in the code blocks like ioon instead of "io on" for example
Actually, I noticed that and checked my original that I submitted to ssc. The code there is fine, I think it has some thing to do with the process used to prepare it for publication.
September 22, 2009 at 8:43 am
Lynn Pettis (9/22/2009)
Silverfox (9/22/2009)
I like the article but just a minor point, the code doesnt work, maybe it is the way it came across when formatted.You are missing a space, in your set statistics statements
in the code blocks like ioon instead of "io on" for example
Actually, I noticed that and checked my original that I submitted to ssc. The code there is fine, I think it has some thing to do with the process used to prepare it for publication.
np, maybe something you can nag steve about :-P, nice article btw, going to take a closer more detailed look when I get a moment.
--------------------------------------------------------------------------------------
[highlight]Recommended Articles on How to help us help you and[/highlight]
[highlight]solve commonly asked questions[/highlight]
Forum Etiquette: How to post data/code on a forum to get the best help by Jeff Moden[/url]
Managing Transaction Logs by Gail Shaw[/url]
How to post Performance problems by Gail Shaw[/url]
Help, my database is corrupt. Now what? by Gail Shaw[/url]
September 22, 2009 at 8:44 am
dbishop (9/22/2009)
I am a bit perplexed by the article. I buiul a tally table in each of my databases a couple of years ago. Static table, 100K rows, PK. When I initially built it I really did not care if it took 150ms or 1.3 seconds, because it was a one-time thing. A tally table is so useful, it should be part of the SQL installation and be included when a new database is created.An article on how to use a tally table would have been a lot better than how to create one. Likle it was said, there are a lot of articles that provide a 5-line code solution on how to create it all over the place. No reason to be creating one on the fly every time you need it.
Curious, what happens if you find yourself in a shop where you aren't allowed to create a static tally table, what are you going to do? This is a problem we see quite frequently on the forums, as Luke mentions above, when a tally table solution is suggested.
In addition, this article may make a little more sense if you take to the time to read the article I also referenced that was published in the May 2009 edition of SQL Server Magazine. This article is in response to that article, demonstrating that there is a better way to generate a dynamic tally table.
September 22, 2009 at 8:46 am
dbishop. See the reply to my first post above regarding the times where a static tally table is not possible.
The article is not about creating a tally table but creating a dynamic equivalent and on those merits it's great.
If you can create a static one then marvelous. If not and you need a dynamic way of achieving the same results each time you need to then performance is everything.
[font="Arial"]_______________________________________________________
Change is inevitable... Except from a vending machine.[/font]
Viewing 15 posts - 1 through 15 (of 159 total)
You must be logged in to reply to this topic. Login to reply