November 20, 2012 at 4:24 am
Hi,
Is it mandatory to have a data warehouse in dimensional model to build a SSAS cubes above?
I'm trying to see if there is any scope to address few reporting needs without the need to build a traditional Data Warehouse first and then build cubes on top of the data warehouse, instead build cubes to do Financial monthly aggregated reporting needs sourcing from backup copy of the Transaction systems.
Appreciate, if somebody can point me to right direction, I'm still learning and hence this silly query.
PS: I have posted same query in Analysis Services Forum assuming this query should have been posted here and not there. Is this okay?
Regards,
KK
November 20, 2012 at 6:04 am
Please don't post the same question in multiple forums; it just splinters the responses.
See my reply at: http://www.sqlservercentral.com/Forums/Topic1386751-17-1.aspx
Rob
November 21, 2012 at 1:54 am
Apologies for this duplicate post. I would like to use this forum for this query.
Many Thanks for your response. Myself coming from traditional data warehouse developing background, believe in building ODS>>DWH>>Report>>Cubes. But due to short time to deliver few Summarized reports I was trying to draw a solution on paper to build cubes. In longer term do you think the traditional approach gives more benefits and is a right approach?
Using the cube approach, I'm not sure if I can consolidate data from heterogeneous sources and build cubes.
Lets say Employee dimension. If I have 3 sources having 3 Employee tables, when building a cube can I build one customer dimension? However, if I take traditional approach I can achieve this.
Regards,
KK
November 21, 2012 at 7:19 am
kamal_ece (11/21/2012)
Lets say Employee dimension. If I have 3 sources having 3 Employee tables, when building a cube can I build one customer dimension? However, if I take traditional approach I can achieve this.
Under the traditional DW approach, your ETL would bring the three sources into a common table and transform the source data so that you have common dimension table.
If you build your cube straight from the relational source data, you should be able to create a SELECT or a View that either JOINs or UNIONs your three data sources into one for an employee dimension.
Again the caveat that I've only built cubes against a dimensional model DW.
Good luck,
Rob
November 24, 2012 at 5:47 am
Many thanks for your valuable inputs. The final solution we are trying to take to address the current reporting solution is to consider the current backup copies as the ODS and build a data Warehouse. By taking this approach the current reporting system is still intact and a new DWH in parallel will be developed. Going forward, new reports to be build on the data warehouse.Slowly decommission the existing reports accessing the backup copies. Interesting thing which I found recently was there is MI reporting on live source systems. Hence, the data warehouse approach. Hopefully, I am on right track.
Building Cubes solution is in consideration but not immediately but may take 6 months later. I feel personally cubes are way to go forward. Having Cubes built for individual departments for MI reporting is a good solution.
Regards,
Kamal
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply