temporal extensions to SQL

I have heard that temporal extensions to ansi/iso standard SQL have been approved and will probably be published this year. I am just wondering how this relates to dimensional modeling.

DBMS vendors seem to be integrating temporal features into their engines as teradata have done in their latest release and ibm are planning for their next release.

Might this help star schema performance ?

Databases have had temporal features for some time now. Oracle has had temporal functions for as long as I can remember that mimic what the new standard proposes, Netezza supports interval data types (in a non-standard manner), etc...

It will certainly make time manipulations and filtering easier to code and make such code more transportable, but I doubt it would have much effect on performance... although I am sure someone can invent a particular use case that would demonstrate some performance advantage, but overall, probably not.

Yes temporal extensions to ansi/iso standard SQL have been approved but are not yet published. These extensions go much further than interval data types as they include support for "business time" and "system time" which together create bitemporal structures.

And yes DMBS vendors are integrating these types of temporal features into their DBMS engines (with varying levels of conformance to the approved ansi/iso extensions). Teradata has already done this and IBM has gone GA with this functionality on their mainframe platform (other platforms to follow in the next year).

As I just mentioned in a post under the "Bi Temporal - As At reporting" topic these extensions are oriented toward a 4 timestamp, penalty on insert approach.

The integration of support for this approach into DBMS engines should go a long way toward addressing bitemporal query performance (as well as complexity, integrity issues etc.)

I do not think integration of these extensions into dbms engines will have a significant impact on star schema performance or on 3 timestamp bitemporal approaches which are sometimes used to augment/stylize fact or dimension tables.