At 05:58 PM 11/21/2002 +0000, Chris Bland wrote:

>The data is loaded in, and is read only after that. We can't predict with
>any certainty which columns will be drilled into, although we are performing
>some analysis on the columns which will allow us to determine which ones
>will benefit most from indexing. The number of columns is arbitrary - some
>users will have very wide files, e.g. 250+ attributes.
>
> >From what I have gleaned, sparse domains are questionable, with selectivity
>being the main deciding factor.
>
>Nevertheless, it would be useful not to have to worry about this maximum
>limit, just applying indexes as we deem appropriate.
>
>Is it there for a particular reason? Is it possible to make it go away...

In fact, the number of indexes that can be created is a function
of the page size and the length of the index name. Larger page sizes
with shorter names yield more indexes. Do create indexes after the
data is stored. For a read-only application, indexes with lots of
duplicates are not great problem - though they would be on read/write
data.

Regards,

Ann
www.ibphoenix.com
We have answers.