 | |
cube strategy and virtual cube
microsoft.public.sqlserver.olap microsoft.public.sqlserver.olap
 |
| | Thread Tools | Display Modes |
#1
 
| |||
| |||
|
cube strategy and virtual cube -
06-21-2005
, 01:01 PM
#2
| |||
| |||
|
|
Hi all In my Fact table I have arrond 8000 million rows and since each cube supports upto 2147,483,647 rows in fact table size. So I created 4 cubes each contain arround 2000 million rows. Is this strategy proper? All this cubes has same measures. So How can I create virtual cube from all cubes which has only one measure which shows data for all cubes. for example I have measure data on each cube. So in virtual cube it shows measures like data , data1 , data2 and data3 for diffrent cubes but I want to show all measures under only one measure Data since all cubes are same only duration of data changed. How can I do it in virtual cube? Thanks Am |
#3
| |||
| |||
|
|
For a cube of that size to get any real performance will need partitioning. Having a single fact table of 2B rows will result in a considerable performance hit if you don't hit an aggregate. It will also take considerable time to process. Both of which is addressed (along with other issues) by using partitioning. -- Dave Wickert [MSFT] dwickert (AT) online (DOT) microsoft.com Program Manager BI SystemsTeam SQL BI Product Unit (Analysis Services) -- This posting is provided "AS IS" with no warranties, and confers no rights. "AM" <anonymous (AT) developersdex (DOT) com> wrote in message news:uhHHUrodFHA.2420 (AT) TK2MSFTNGP12 (DOT) phx.gbl... Hi all In my Fact table I have arrond 8000 million rows and since each cube supports upto 2147,483,647 rows in fact table size. So I created 4 cubes each contain arround 2000 million rows. Is this strategy proper? All this cubes has same measures. So How can I create virtual cube from all cubes which has only one measure which shows data for all cubes. for example I have measure data on each cube. So in virtual cube it shows measures like data , data1 , data2 and data3 for diffrent cubes but I want to show all measures under only one measure Data since all cubes are same only duration of data changed. How can I do it in virtual cube? Thanks Am |
#4
| |||
| |||
|
|
Ok I will partition it. But can I use one cube for data about 8000 million and one fact table. -- Thanks "Dave Wickert [MSFT]" <dwickert (AT) online (DOT) microsoft.com> wrote in message news:upo3fsrdFHA.3492 (AT) TK2MSFTNGP14 (DOT) phx.gbl... For a cube of that size to get any real performance will need partitioning. Having a single fact table of 2B rows will result in a considerable performance hit if you don't hit an aggregate. It will also take considerable time to process. Both of which is addressed (along with other issues) by using partitioning. -- Dave Wickert [MSFT] dwickert (AT) online (DOT) microsoft.com Program Manager BI SystemsTeam SQL BI Product Unit (Analysis Services) -- This posting is provided "AS IS" with no warranties, and confers no rights. "AM" <anonymous (AT) developersdex (DOT) com> wrote in message news:uhHHUrodFHA.2420 (AT) TK2MSFTNGP12 (DOT) phx.gbl... Hi all In my Fact table I have arrond 8000 million rows and since each cube supports upto 2147,483,647 rows in fact table size. So I created 4 cubes each contain arround 2000 million rows. Is this strategy proper? All this cubes has same measures. So How can I create virtual cube from all cubes which has only one measure which shows data for all cubes. for example I have measure data on each cube. So in virtual cube it shows measures like data , data1 , data2 and data3 for diffrent cubes but I want to show all measures under only one measure Data since all cubes are same only duration of data changed. How can I do it in virtual cube? Thanks Am |
#5
| |||
| |||
|
|
As I said there is no technical limit. However, effectively have 8BILLION rows in a single file (which is how a partition is structured) will perform slowly compared to using partitioning (for most common queries). -- Dave Wickert [MSFT] dwickert (AT) online (DOT) microsoft.com Program Manager BI SystemsTeam SQL BI Product Unit (Analysis Services) -- This posting is provided "AS IS" with no warranties, and confers no rights. "AM" <anonymous (AT) developersdex (DOT) com> wrote in message news:uCYXwW2dFHA.688 (AT) TK2MSFTNGP14 (DOT) phx.gbl... Ok I will partition it. But can I use one cube for data about 8000 million and one fact table. -- Thanks "Dave Wickert [MSFT]" <dwickert (AT) online (DOT) microsoft.com> wrote in message news:upo3fsrdFHA.3492 (AT) TK2MSFTNGP14 (DOT) phx.gbl... For a cube of that size to get any real performance will need partitioning. Having a single fact table of 2B rows will result in a considerable performance hit if you don't hit an aggregate. It will also take considerable time to process. Both of which is addressed (along with other issues) by using partitioning. -- Dave Wickert [MSFT] dwickert (AT) online (DOT) microsoft.com Program Manager BI SystemsTeam SQL BI Product Unit (Analysis Services) -- This posting is provided "AS IS" with no warranties, and confers no rights. "AM" <anonymous (AT) developersdex (DOT) com> wrote in message news:uhHHUrodFHA.2420 (AT) TK2MSFTNGP12 (DOT) phx.gbl... Hi all In my Fact table I have arrond 8000 million rows and since each cube supports upto 2147,483,647 rows in fact table size. So I created 4 cubes each contain arround 2000 million rows. Is this strategy proper? All this cubes has same measures. So How can I create virtual cube from all cubes which has only one measure which shows data for all cubes. for example I have measure data on each cube. So in virtual cube it shows measures like data , data1 , data2 and data3 for diffrent cubes but I want to show all measures under only one measure Data since all cubes are same only duration of data changed. How can I do it in virtual cube? Thanks Am |
|
« Previous Thread
|
Next Thread »
| Thread Tools | |
| Display Modes | |
Linear Mode
| |
Powered by vBulletin Version 3.5.3
Copyright ©2000 - 2012, Jelsoft Enterprises Ltd.
Copyright ©2000 - 2012, Jelsoft Enterprises Ltd.