Re: Why DISTINCT is costly?

Dan Koren wrote:

>
>
> And how else would one be able to
> eliminate duplicates without sort
> unique?

It should be able to be done with an index. Whether the optimizer will use the index to retrieve unique values without a SORT UNIQUE is another question. The few rudimentary tests I've run show that it doesn't. It still greatly improves the performance though since the index is scanned and not the table.

-- 
To reply by email remove "_nospam"