I am not really sure wether I MUST use it or not, but I think it is a
nice feature to be utilized.
What I want is a long running producing up to hundred of thousands
results which have to undergo a computationally expensive
postprocessing step which needs to be execeuted in parallel. By using
pipelining I would not need to worry about partitioning the query and
thus distributing the workload.
At least as far as I understand the pipelining feature...
tschak