Svein Erling Tysvær wrote:

> Maybe UNION could help (speedwise)?
>
>

After Anderson's post, I thought about replacing with a union. At least
it could use the index I think would help. Perhaps a Stored procedure
might be even better.

It isn't so much the speed of execution as one of fetching.

> (Actually, I've never tried using DISTINCT with UNION statements, just hope it works - for all I know, it might even be redundant)
>
> For this to be worth trying, there must be considerably more rows in Deal than there are records in Discussion or Attachment with CreateDate >= '3/1/2008'.
>
>

I think the problem is more with fetches from unnecessary Discussion
rows (perhaps not to the client, that confuses me somewhat, but
certainly ibexpert shows many row fetches that I know are unnecessary
and do not make it to the final result).

This degradation may be exaggerated because A) a Discussion row is large
in size B) this is a remote application in a 3 tier environment
communicating over the internet.

Thanks Set,
Mitch