RE: Q: redo bytes statistics is it sum of all members or not ?

From: Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk>
Date: Wed, 25 Jun 2014 12:43:35 +0000
Message-ID: <CE70217733273F49A8A162EE074F64D901DFD271_at_exmbx05.thus.corp>

That would give you the total write I/O to log files due to online redo log writes.

Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
_at_jloracle

---

From: amihay gonen [agonenil_at_gmail.com] Sent: 25 June 2014 13:35
To: Jonathan Lewis
Cc: ORACLE-L
Subject: Re: Q: redo bytes statistics is it sum of all members or not ?

ok , according to this , oracle doesn't have a clear metric how many redo bytes where actually written to disk .

what i can do it to multiple redo blocks written* blk size*member_count . ( making sure that all members are valid and not stale).

On Wed, Jun 25, 2014 at 3:07 PM, Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk<mailto:jonathan_at_jlcomp.demon.co.uk>> wrote:

Physical Write Bytes is the "logical" bytes for blocks written by DBWR.

Physical write total bytes is EVERYTHING else - including every copy of the redo logs written by LGWR, control file writes and data files writes from CKPT, control file writes from MMON, archived redo log writes from ARC0, direct path writes and local writes from foreground sessions etc...

Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
_at_jloracle

---

From: amihay gonen [agonenil_at_gmail.com<mailto:agonenil_at_gmail.com>] Sent: 25 June 2014 12:48
To: Jonathan Lewis
Cc: ORACLE-L

Subject: Re: Q: redo bytes statistics is it sum of all members or not ?

thanks for the quick response.

I still don't understand . that stat : physical write total bytes according to oracle doc is "Total size in bytes of all disk writes for the database instance including application activity, backup and recovery, and other utilities. The difference between this value and "physical write bytes" gives the total write size in bytes by non-application workload."

1. is this metric means all writes (include dbwr writes ) ?
2. if I reduce physical write total bytes-physical write bytes it will be equal to redo size+ redo wastage ? is lgwr part of application work load .
3. is the answer to 1 is true , the mulitple "physical write total bytes" * redo_memebers_number , won't give correct answer.

thanks for your time,
amihay

On Wed, Jun 25, 2014 at 1:39 PM, Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk<mailto:jonathan_at_jlcomp.demon.co.uk>> wrote:

You're correct.
You could think of it as a "logical" measure of redo generated and written.

When you look at the stat physical write total bytes", that's when you have to multiple by the number of copies before adding to the total, and don't forget that "redo blocks written" is derived from "redo size" + "redo wastage".

Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
_at_jloracle

---

From: oracle-l-bounce_at_freelists.org<mailto:oracle-l-bounce_at_freelists.org> [oracle-l-bounce_at_freelists.org<mailto:oracle-l-bounce_at_freelists.org>] on behalf of amihay gonen [agonenil_at_gmail.com<mailto:agonenil_at_gmail.com>] Sent: 25 June 2014 11:09
To: ORACLE-L
Subject: Re: Q: redo bytes statistics is it sum of all members or not ?

some question for "redo size" statistic as well

thanks .

On Wed, Jun 25, 2014 at 12:51 PM, amihay gonen <agonenil_at_gmail.com<mailto:agonenil_at_gmail.com>> wrote: Hi DBA's - quick question

According to my tests the statistic "redo blocks written" is not a sum of redo block written to all redo memebers.

I've run test with instance with 2 members and then with 3 members and this metric didn't change.

but just to make sure I didn't missed something :)

I wander if anyone has test it and can approved this conclusion .

thanks ,
amihay

--
http://www.freelists.org/webpage/oracle-l

Received on Wed Jun 25 2014 - 14:43:35 CEST