Virtual Cubes

Post Reply
guntupalli
Posts: 18
Joined: Mon May 16, 2011 12:15 pm
OLAP Product: tm1
Version: 10.2
Excel Version: 2010

Virtual Cubes

Post by guntupalli » Wed Jul 27, 2011 12:19 pm

Hi ,

can anyone explain what is Virtual Cubes ?
difference between cube and Virtual Cubes?
how can we create and it works?

i know few points about it :
Virtual Cubes don't store the data and driven by rules,
Virtual Cubes used for instead of browsing the cube directly and summery data .

Thanks in Advance
Best Regards,
Guntupalli

User avatar
qml
MVP
Posts: 1074
Joined: Mon Feb 01, 2010 1:01 pm
OLAP Product: TM1 / Planning Analytics
Version: 2.0.6 and all previous
Excel Version: 2007 - 2016
Location: London, UK, Europe

Re: Virtual Cubes

Post by qml » Wed Jul 27, 2011 1:12 pm

You should probably ask these questions to the person that used this term.

In TM1 there is no concept of virtual cubes, but I can imagine (and your points seem to confirm it) that this could be a convenient shorthand meaning a regular cube with no data stored in it and all data pulled/calculated by rules. This is not terminology that I know to be universally accepted/used within the TM1 community, however.
Kamil Arendt

guntupalli
Posts: 18
Joined: Mon May 16, 2011 12:15 pm
OLAP Product: tm1
Version: 10.2
Excel Version: 2010

Re: Virtual Cubes

Post by guntupalli » Wed Jul 27, 2011 1:46 pm

Thanks for reply ,
In TM1 there is no concept of virtual cubes
i found one topic about virtual cube in below url.

http://www-304.ibm.com/jct03001c/servic ... Code=P6502.

any information about my questions , plz share..

thank you .
Best Regards,
Guntupalli

User avatar
qml
MVP
Posts: 1074
Joined: Mon Feb 01, 2010 1:01 pm
OLAP Product: TM1 / Planning Analytics
Version: 2.0.6 and all previous
Excel Version: 2007 - 2016
Location: London, UK, Europe

Re: Virtual Cubes

Post by qml » Wed Jul 27, 2011 2:04 pm

I'm sure that someone charging $3,500.00 plus tax is able to provide much more information on this subject than I am. Everything I know about virtual cubes (that don't exist) in TM1 is in my post above. But at least that's free of charge.
Kamil Arendt

David Usherwood
Site Admin
Posts: 1377
Joined: Wed May 28, 2008 9:09 am

Re: Virtual Cubes

Post by David Usherwood » Wed Jul 27, 2011 2:32 pm

Those of us who have done the TM1 certifications know that IBM have their own strange language to describe terms and concepts which the 'real' TM1 world doesn't recognise until.... we do a TM1 cert.

User avatar
qml
MVP
Posts: 1074
Joined: Mon Feb 01, 2010 1:01 pm
OLAP Product: TM1 / Planning Analytics
Version: 2.0.6 and all previous
Excel Version: 2007 - 2016
Location: London, UK, Europe

Re: Virtual Cubes

Post by qml » Wed Jul 27, 2011 2:39 pm

I've done TM1 Developer cert and I can't recall anything about "virtual cubes". Is that why I didn't get 100%? :mrgreen:
Kamil Arendt

applix_mc
Posts: 16
Joined: Mon Jun 02, 2008 3:11 pm

Re: Virtual Cubes

Post by applix_mc » Wed Jul 27, 2011 5:35 pm

I have some old notes regarding how to create a virtual cube - this may provide some insight. I personally never done this but had this old doc hidden in some archives
Attachments
virtualcube.doc
Virtual Cube concept
(26 KiB) Downloaded 350 times

User avatar
Martin Ryan
Site Admin
Posts: 1957
Joined: Sat May 10, 2008 9:08 am
OLAP Product: TM1
Version: 10.1
Excel Version: 2010
Location: Wellington, New Zealand
Contact:

Re: Virtual Cubes

Post by Martin Ryan » Thu Jul 28, 2011 1:37 am

To paraphrase applix_mc's diagram and David's reference to the IBM certifications, a virtual cube is a fully rule calculated one. There is no data stored, just rules pulling data from other cubes.

To my recollection I have never created a 100% rule calculated (i.e. "virtual") cube. Ever. Because why on earth would I just pull the data from another cube or cubes into another cube, when I already obviously already have that data available in the original cubes? And if I needed to do a calculation on that data it's almost guaranteed that it would make more sense to do that calculation in one of the source cubes.

So it's (IMHO) a dumb concept that you can forget once you've completed the certification.

Martin
Please do not send technical questions via private message or email. Post them in the forum where you'll probably get a faster reply, and everyone can benefit from the answers.
Jodi Ryan Family Lawyer

User avatar
Alan Kirk
Site Admin
Posts: 5996
Joined: Sun May 11, 2008 2:30 am
OLAP Product: TM1
Version: PA2 Classic (PAW-free zone)
Excel Version: 2010 and 2016
Location: Sydney, Australia
Contact:

Re: Virtual Cubes

Post by Alan Kirk » Thu Jul 28, 2011 2:09 am

Martin Ryan wrote:To paraphrase applix_mc's diagram and David's reference to the IBM certifications, a virtual cube is a fully rule calculated one. There is no data stored, just rules pulling data from other cubes.

To my recollection I have never created a 100% rule calculated (i.e. "virtual") cube. Ever. Because why on earth would I just pull the data from another cube or cubes into another cube, when I already obviously already have that data available in the original cubes? And if I needed to do a calculation on that data it's almost guaranteed that it would make more sense to do that calculation in one of the source cubes.

So it's (IMHO) a dumb concept that you can forget once you've completed the certification.
I have to admit that I have two. They're part of a combined costing and revenue model that I inherited. The model could have been done more cleanly but I have to admit that even if I was redoing it from scratch I'd still have the two rules-based cubes. They're currently fed from 6 other cubes (though this could have been reduced to four). Inputs come from different parts of the business, which have access to different input cubes. The cost and revenue outputs are derived from many of the inputs, but not in such a way that they tie in a sensible manner to any one of the specific input cubes; they're merely the end result of the various inputs, as well as intermediate calculations which drive various statistical values. Put simply, it's a lot cleaner to have a separation between the "input" cubes and the "output" cubes in a fairly complex collaborative model like this one.

I will, though, happily concede that such a model is the exception rather than the rule.

However I've never regarded the output cubes as being "virtual"; they're no different to any other cubes, save for the fact that they don't have direct inputs. (But could if I needed them to.)
"To them, equipment failure is terrifying. To me, it’s 'Tuesday.' "
-----------
Before posting, please check the documentation, the FAQ, the Search function and FOR THE LOVE OF GLUB the Request Guidelines.

lotsaram
MVP
Posts: 3317
Joined: Fri Mar 13, 2009 11:14 am
OLAP Product: TableManager1
Version: PA 2.0.x
Excel Version: Office 365
Location: Switzerland

Re: Virtual Cubes

Post by lotsaram » Thu Jul 28, 2011 7:01 am

Martin Ryan wrote:To paraphrase applix_mc's diagram and David's reference to the IBM certifications, a virtual cube is a fully rule calculated one. There is no data stored, just rules pulling data from other cubes.

To my recollection I have never created a 100% rule calculated (i.e. "virtual") cube. Ever. Because why on earth would I just pull the data from another cube or cubes into another cube, when I already obviously already have that data available in the original cubes? And if I needed to do a calculation on that data it's almost guaranteed that it would make more sense to do that calculation in one of the source cubes.

So it's (IMHO) a dumb concept that you can forget once you've completed the certification.

Martin
I never heard of a virtual cube either and I have done all the certs. A cube is a cube whether is only has rules or whether it has input data.

I have used them though and wouldn't think a rule only cube to be that uncommon. I can think of a few cases where they have been handy.
1/ Reconciliation between cubes that *should* tally but have separate data sources or rollup structures.
2/ Recolciliation between a retail analytics cube at department level that had been enriched with additional dimensions (size, color, brand, license, vendor ....) which were only available in the source system as SKU level attributes to built the dept cube by accumulating from the the SKU level source cube. (Had tiered structure for performance since most of the time users only needed dept level not SKU level reports)
3/ Source system saw the CoA as flat and due to the way adjustments/corrections/eliminations were journaled if you just pulled the posting accounts and rolled up by hierarchy in a TM1 cube you might not reconcile exactly back to the source. Therefore had a cube that mirrored the source (that is flat CoA) but mirrored a reporting cube fed by rule with CoA that users could drill up and down which was more user friendly. (Hey don't ask me to comment on the quality or otherwise of the GL.)
4/ When creating a 2nd "virtual" cube to check for overfeeding of a cube of interest

Post Reply