| BSO | ASO | |
| Build dimensions | Y | Y |
| Load data | Y | Y |
| Write back to level 0 | Y | Y |
| Retrieve data | Y | Y |
| Partitioning / Sharing Data | Y | Y (target or source in 11x) |
Key differences ...
| BSO | ASO | |
| Unlimited dimensions and members | N | Y |
| Write back to any level | Y | N |
| Advanced calculation engine | Y | N |
| Calc scripts vs. MDX formulas | Y | N |
| Partitioning / Sharing Data | Y | Limited – pre-11x |
In short, use BSO when you need to write back to any level or perform complex calculations. Use ASO for the larger, aggregation focused databases with many dimensions and many members.
In the 11 world where I can now make an ASO database the source of a partition, I can take advantage of the BSO strengths (write back to any level, powerful calculation engine) and then source this information to a consolidated ASO database that maybe has the volumes of detail from other sources. (Note - the new Hyperion Profitability and Cost Management solution uses this model: BSO for allocation calcs and loads to an ASO cube for reporting).
Check out the DBAG for detailed comparison tables on ASO vs. BSO.
7 comments:
Hi
Can I have ASO and BSO Cubes on one server?
Regards
Jayanth
@jayanth - yes you can. The server specifications are the limitaiton here.
I am not agree with the UNLIMITED DIMENSION in ASO cubes. That is not true in practice...
I am not agree with the UNLIMIED DIMENSION definition for ASO cubes. That is not true in practice...
Should there be another persuasive post you can share next time, I’ll be surely waiting for it. Sarah Canning
So luck to come across your excellent blog. Your blog brings me a great deal of fun .. Good luck with the sit e. Michael
What a blog post!! Very informative and also easy to understand. Looking for more such comments!! Do you have a facebook? I recommended it on digg. The only thing that it’s missing is a bit of new design. etf rotation
Post a Comment