I am nowadays trying to study about query optimization in replicated distributed database system. But I got myself confused about joining order of the relations.
If a query execution path is
[15E,4F,6I,8B,14D,11G,16J,9H,6A,13C] or [1I,7D,8G,6A,15C,12B,7E,12H,9J,3F]
where every element is composed of a site number where the relation is resided. Then, does the result vary from each other?
Thanks