Oracle Open World Essbase Roadmap Session

For me, one of the highlights of Oracle Open World is the ability to attend a number of roadmap sessions where Oracle tells you what they plan to work on over the next year.  Of course, these sessions are always prefaced with the standard Safe Harbor statements that the contents are not officially part of the product until they ship, that they may not ever ship as part of the product, don’t make any decisions based on the content as the features discussed are subject to change, etc..  The items on the list, however, do show the continued investment, and thus continued commitment, that Oracle has to Essbase.

This year, the items mentioned as areas of concentration over the next year show continued work on Essbase scalability, dependability, performance and integration with the Oracle Fusion strategy.  From my quickly and messily scribbled notes, some of the items on the list include:

• 6 new calc functions
• Parallel Fix
• Java API performance improvements
• APS/Essbase custom aggregates and lists
• Dimension properties
• MDX improvements
• Fusion Middleware security integration
• Dimension build performance increases
• Calc runtime parameters
• Resolve naming restrictions

My favorite items on the list are the Java API performance improvements and the Essbase custom aggregates.  We use exclusively the Java API inside Dodeca and any performance improvements can only be a good thing.  Personally, I think that grid retrievals are quite fast in the Java API already.  In fact, we see retrievals in Dodeca that are faster than the classic Excel add-in.  That being said, many of the APIs that query member information are pitifully slow plus they consume an unnecessarily large number of ports.  

On the other hand, I can only imagine the functionality that Essbase custom aggregates could bring to the product.  From the sounds of it, it will provide a way to define custom members that simply provide a total of values from other selected members in a dimension.   For example, I can envision a use-case where I could define “Tim’s Region” as the total of Michigan, California and Oregon.  This personal member would then be available for retrieve/zoom/pivot and other Essbase operations.   Until we actually see it, we will not know for sure what it can do.  I do hope, however, it is exposed via the Java API so we can develop some cool solutions around them.

Which of these features would you like to see in the product?  What features would you like to see that are *not* on this list?