Like I said:
"So, my question is one of *general usage hence the lack of an example*. Is
there some general feature, (compiled code for example) that can be applied
to make SQL with embedded user-defined functions run faster?"
But thanks for your reply. :-)
Mike
On Fri, Jul 31, 2009 at 9:45 AM, Javier Montani <jmont...@gmail.com> wrote:
> Can you paste the trace / execution plan?
>
> 2009/7/31 kevin <majun...@hotmail.com>
>
>
>>
>> Don't use function against table columns, just against your Constant,
>> for example:
>>
>> don't do: where to_char(column_name, 'mmddyyyy') = '07312009'
>> just to: where column_name = to_date('07312009', 'mmddyyyy')
>>
>> On Jun 26, 8:06 am, "Rob Wolfe" <rob.wo...@oraclegeeks.com> wrote:
>> > One thing that I have noted (this is purely anecdotal, I have no
>> numbers)
>> > with functions that I write is that if I package up ones that are
>> > temporally close to each other it seems to be somewhat quicker than if
>> > they are standalone functions. I have no real rationale as to why this
>> > would be the case unless it is due to the entire package getting cached
>> > but that is a complete guess.
>> >
>> > David might have some hard numbers to back this up or refute it.
>> >
>> > One other thing that i have noticed is that once dblinks and user
>> > functions get together performance goes straight to hell unless you are
>> > VERY careful with what you are doing. I am not sure if that is due to
>> > horrible configuration errors that we have made (possible), if I just
>> suck
>> > (also possible) or if it is something inherent in the beast.
>> >
>> > I take advantage of the parallel_enable hint/option/keyword whenever I
>> > can as well.
>> >
>> > well, that is what little wisdom I can offer on this one. I think i
>> shall
>> > go have a drink in memory of Michael Jackson and Farah Fawcett... 47
>> > suddenly feels much older than it did yesterday.
>> >
>> > On Thu, June 25, 2009 13:42, Michael Moore wrote:
>> >
>> > > A co-worker has the problem where he has created several functions
>> that he
>> > > uses in SQL statements. His observation if that these functions, even
>> > > though
>> > > light weight, really slows down his SQL.
>> > > We have speculated that this user-defined function is causing context
>> > > switches which may be (at least in part) the cause of the slowness.
>> >
>> > > So, my question is one of general usage hence the lack of an example.
>> Is
>> > > there some general feature, (compiled code for example) that can be
>> > > applied
>> > > to make SQL with embedded user-defined functions run faster?
>> >
>> > > Mike
>> >
>> >
>>
>>
>
> >
>
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google
Groups "Oracle PL/SQL" group.
To post to this group, send email to Oracle-PLSQL@googlegroups.com
To unsubscribe from this group, send email to
oracle-plsql-unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/Oracle-PLSQL?hl=en
-~----------~----~----~----~------~----~------~--~---
