Star Schema for MPP databases

Have anyone designed Star Schema for MPP databases?  If so, can someone point me to an article or paper that covers this topics? Thanks, ..rao

Here's the closest I've found, http://www.slideshare.net/pepeborja/teradata-aggregate-join-indices-and-dimensional-models-4860226?from=share_email

It depends on which MPP you are working with.

Netezza handles star schema extremely well.  The only major consideration is when you have one very large dimension (50M+ rows).  Typically you would distribute the fact on that dimension key, to achieve a co-located join on that dimension.  Smaller dimension are usually not a concern, and anything under 1M rows isn't any bother at all.  Dimensions should be distributed on their surrogate PK.

With Teradata you need to build 'join indexes' which are essentially pre-joined materialized views.  (Buy a lot more disk)