Re: Performance problems with distributed queries with table joins.

Since the optimizer knows nothing about the remote tables, it will be inefficient running distributed queries. I remember working with some large distributed queries in V6. For that application I eventually went to dynamically creating views of the required tables which limited the tables based on criteria and then executed the joins between the views. The trick was to use "complex" views which did grouping so that the view results were created first (as temp sorted data) and then sort/merged.

V.

-- 
#include usual disclamers, warrenties, and other lies