phys reads vs blocks read - statspack's segment stats

From: Boris Dali <boris_dali_at_yahoo.ca>
Date: Wed, 31 Dec 2003 07:49:35 -0800
Message-ID: <F001.005DB5D4.20031231074935@fatcity.com>

I am reviewing segment statistics taken in a 15 min snapshot (from stats$seg_stat) for a specific table, which resides in 2 datafiles. The stats for datafiles
(stats$filestatxs) are as follows:

file# blocks read physical_reads
----- ----------- --------------

11          8,171            960
12          6,149            387

while the table stats are:

lio db block changes phyrds

------ ---------------- -------
42,064                0  10,821

One would expect stats$seg_stat.physical_reads for specific segment/snapshot to be <=
stats$filestat.physrds for the datafiles, segment resides in, which doesn't seem to be the case.

Number of reads from the table somewhat makes sense if compared against number of blocks read from the datafiles, but not against number of reads from these datafiles.

I know v$filestat timing info is notoriously unreliable, but I thought i/o (number and blocks) stuff can be trusted to.

Am I missing something obvious?

Thanks,
Boris Dali.

Post your free ad now! http://personals.yahoo.ca

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.net
-- 
Author: Boris Dali
  INET: boris_dali_at_yahoo.ca

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L


(or the name of mailing list you want to be removed from).  You may

also send the HELP command for other information (like subscribing).

Received on Wed Dec 31 2003 - 09:49:35 CST