Ron Mayer wrote:
>> Or do you mean being able to perform parts of the query plan fully in
>> parallel? If this, then one would need a lot more than ParallelSort...
>>
>
> I wouldn't recommend that - it seems like a Hard Problem.
>
>
Isn't it the case that the implicit unions from processing partitioned
data provides a
more-or-less-ideal opportunity here?
I certainly have sympathy for parallelising expensive queries to bring
the best response
time down, even if the average under full load goes up slightly, since
any implied locks
(including pinning of read-ahead ages) will be released sooner.
And when load is light, users who are online get more of the hardware
they paid for.
James