Re: explain plan, can you explain this?

well, unless I have run into 3 really strange instances, it looks like
cumulative values to me.  Unless you can explain why the phyrds and phywrts
are higher for each snap than the previous one for every file in my
instance.

On Jan 11, 2008 7:28 AM, Yasin Baskan <yasin.baskan@xxxxxxxxxxxxxxxx> wrote:

>  File stats are also reported as delta.
>
>
>
> ...
>
> sum(e.phyrds  - nvl(b.phyrds,  0))))      atpr,
>
> ...
>
> from dba_hist_filestatxs e,
>
>
>
>            dba_hist_filestatxs b
>
>      where b.snap_id(+)         = :bid
>
>
>
>    and e.snap_id            = :eid
>
>
>
> I think it would defeat all the purpose of snapshots if it were
> cumulative.
>
>
>
> *From:* Andrew Kerber [mailto:andrew.kerber@xxxxxxxxx]
> *Sent:* Friday, January 11, 2008 3:24 PM
> *To:* Yasin Baskan
> *Cc:* john.kanagaraj@xxxxxxxxx; dannorris@xxxxxxxxxxxxx; Oracle L
> *Subject:* Re: explain plan, can you explain this?
>
>
>
> Thats from dba_hist_sqlstat.  Try dba_hist_filestatxs
>
> On Jan 11, 2008 3:04 AM, Yasin Baskan <yasin.baskan@xxxxxxxxxxxxxxxx>
> wrote:
>
> Andrew, AWR reports delta values, not cumulative statistics.
>
>
>
> It reports the delta values from the columns of dba_hist_sqlstat like
> buffer_gets_delta.
>
>
>
> Here is a part of an sql run by a 10.2 awrrpt.sql
>
>
>
> ...
>
> sum(disk_reads_delta) dskr, sum(executions_delta) exec,
>
>
>
>                       sum(cpu_time_delta) cput, sum(elapsed_time_delta)
> elap
>
> ...
>
> from dba_hist_sqlstat
>
>                   where dbid
>
>       = :dbid
>
>                     and instance_number = :inst_num
>
>
>
>           and :bid            < snap_id
>
>                     and snap_id
>
>    <= :eid
>
>                group by sql_id)
>
> ...
>
>
>
> *From:* Andrew Kerber [mailto:andrew.kerber@xxxxxxxxx]
> *Sent:* Thursday, January 10, 2008 7:57 PM
> *To:* john.kanagaraj@xxxxxxxxx
> *Cc:* Yasin Baskan; dannorris@xxxxxxxxxxxxx; Oracle L
> *Subject:* Re: explain plan, can you explain this?
>
>
>
> And then, you have to do it again for AWR.  AWR does report cumulative
> statistics.
>
> On Jan 10, 2008 11:36 AM, John Kanagaraj <john.kanagaraj@xxxxxxxxx >
> wrote:
>
> > John, are you sure that statspack reports sql statistics cumulatively?
>
> Yasin and Allen,
>
> You are right: My bad. STATSPACK indeed does report the differences.
> This myth of mine was self-created, probably from early days of
> STATSPACK. Thank you for setting me right. This reminds me once more
> to re-validate everything I know.
>
> Regards,
>
> --
> John Kanagaraj <><
> DB Soft Inc
> http://www.linkedin.com/in/johnkanagaraj
> http://jkanagaraj.wordpress.com (Sorry - not an Oracle blog!)
> ** The opinions and facts contained in this message are entirely mine
> and do not reflect those of my employer or customers **
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>
>
> --
> Andrew W. Kerber
>
> 'If at first you dont succeed, dont take up skydiving.'
>
>
>
>
> --
> Andrew W. Kerber
>
> 'If at first you dont succeed, dont take up skydiving.'
>



-- 
Andrew W. Kerber

'If at first you dont succeed, dont take up skydiving.'

Other related posts: