I dont think your schema design will work for star transformation. I believe the requirement is: - primary key on column of dimension table - bitmap index on SK column on fact table In both cases its single column keys. Given you cant have 2 single column primary keys on your dimension table, it would seem it would not work. In good design there is a single column PK and SK on the dimension and fact table. On Thu, Oct 7, 2010 at 12:25 AM, amonte <ax.mount@xxxxxxxxx> wrote: > Hi > > I am reading 11.1 and 11.2 Data Warehousing Guide and where it explains > about using star transformation with bitmap index in both release it states: > > *A prerequisite of the star transformation is that there be a > single-column bitmap index on every join column of the fact table. These > join columns include all foreign key columns.* > > Is it saying that bitmap index cannot be composite? > > So for example if fact table has FK against dimension such as > > fact.c1 ----> dim1.c1 > fact.c2 ----> dim1.c2 > > We cannot create a bitmap index on fact.c1 and fact.c2? But we can create > two seperate bitmap index each for each fact table column? > > > Alex > > -- Regards, Greg Rahn http://structureddata.org