Re: tkprof interpreatation question

On Mar 6, 11:48 am, "Steve Howard" <stevedhow..._at_gmail.com> wrote:
> On Mar 5, 5:46 pm, "joel garry" <joel-ga..._at_home.com> wrote:
>
>
>
>
>
> > On Mar 5, 1:12 pm, harvinde..._at_gmail.com wrote:
>
> > > Thanks, With this parameter can we only specify the buffer size in
> > > bytes or is there any specific parameter with which we can specify the
> > > number of rows per fetch irrespective of bytes.
>
> > I doubt it, wouldn't that mean the buffer size would always be
> > changing, with all the attendant overhead that implies? Things work
> > so much better when the buffer is fixed and Oracle can just jam in
> > whatever data it has up to that size.
>
> > jg
> > --
> > @home.com is bogus.
> > Quick Find Knowledge Base "rman crosscheck"
> > "Did you mean RMAN crotches" - metalink.
>
> I would agree with this logically, except SQL*PLUS allows you to do
> it, as does JDBC (and probably others)...
>
> set arraysize
> setFetchSize(), etc

You are correct, of course.

Then again, "The effectiveness of setting ARRAYSIZE depends on how well Oracle Database fills network packets and your network latency and throughput. In recent versions of SQL*Plus and Oracle Database, ARRAYSIZE may have little effect. Overlarge sizes can easily take more SQL*Plus memory which may decrease overall performance." I would guess some research might show interesting non-symmetry between raising and lowering buffer sizes.

I won't even begin to address the performance of anything with a J in the name.

I can think of many things one can do that one would not want to do. Yet, they are done.

jg

--
@home.com is bogus.
Wait a minute, _I_ have a J in my name...