Re: Why tot parses <> executions in bstat/estat #s?

declare

    mv varchar2(32);
begin

    for i in 1..10 loop

        select user into mv from dual;
    end loop;
end;
.

Execute this, and you parse

    select user from dual
once, but execute it 10 times.
Parses will go up by 1, executes will go up by 10.

PL/SQL, OCI. and Pro*3gl programs all have the ability to re-execute without reparsing.

--
Jonathan Lewis
Yet another Oracle-related web site:  http://www.jlcomp.demon.co.uk

Practical Oracle 8i:  Building Efficient Databases

Publishers:  Addison-Wesley
See a first review at:
http://www.ixora.com.au/resources/index.htm#practical_8i
More reviews at: http://www.jlcomp.demon.co.uk/book_rev.html



H2oStyle wrote in message <91tkmo$k78$1_at_nnrp1.deja.com>...


>right, if it's not re-parsed (hard parsed) then you'd think it's using

>existing parsed sql.  But then you might also think that would mean

>that "parse count (total)" which includes "parse count (hard)" would

>equal "execute count".  But in all of my bstat/estat reports, they

>don't.  By the way, it seems that the more hard parses you have the

>closer the numbers get to being equal.

>

>Thank,

>Scott

>

>------------------------------

>In article <91thf7$hdl$1_at_nnrp1.deja.com>,

>  steveee_ca_at_my-deja.com wrote:

>> Hi Scott,

>>

>> Also,if an SQL statement is issued that's identical to one that's

>> already in the Shared Pool it doesn't have to be re-parsed.

>>

>> Hth,

>>

>> Steve

>>

>> In article <91tbo6$c4j$1_at_nnrp1.deja.com>,

>>   H2oStyle <sdempsey_at_my-deja.com> wrote:

>> > Anyone know why in the bstat/estat report the number of total parses

>> > doesn't equal the execute count?

>> >

>> > The other related or underlying question was that if it's not a hard

>> > parse, what is it?  Most people say its a soft parse meaning that it

>> > found a shared sql statement in the cache to use.  But then we come

>> > back to the original question.

>> >

>> > I'm wondering if perhaps it might have something related to do with

 the


>> > way the client tries to exec sql or rather the way it passes it to

>> > oracle.

>> >

>> > Thanks,

>> > Scott

>> >

>> > Sent via Deja.com

>> > http://www.deja.com/

>> >

>>

>> Sent via Deja.com

>> http://www.deja.com/

>>

>

>

>Sent via Deja.com

>http://www.deja.com/