Re: Inheritance efficiency

2010/4/26 Vincenzo Romano <vincenzo(dot)romano(at)notorand(dot)it>:
> 2010/4/26 Bruce Momjian <bruce(at)momjian(dot)us>:
>> Vincenzo Romano wrote:
>>> Hi all.
>>>
>>> I'm wondering how efficient the inheritance can be.
>>> I'm using the constraint exclusion feature and for each child table
>>> (maybe but one) I have a proper CHECK constraint.
>>> How efficient can the query planner be in choosing the right child
>>> tables in the case of, say, thousands of them?
>>> Would the selection process behave linearly, logarithmically or what?
>>
>> It is fine for dozens of child tables, but not thousands; it does need
>> improvement.
>
> This sounds like "linear" algorithms. Doesn't it?
>
>>> And now it comes to my mind the same question for partial indexes.
>>> That is, if I had a lot (really a lot) of small partial indexes over a
>>> very large table, how efficient can the query planner be
>>> in selecting the right indexes?
>
> No info about this point (partial indexes)?
> Is also this geared with linear algorithms ?

Should I move to an "enterprise grade" version of PostgreSQL?

--
Vincenzo Romano
NotOrAnd Information Technologies
NON QVIETIS MARIBVS NAVTA PERITVS