Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> Re: Raw devices and redo

Re: Raw devices and redo

From: Tim Gorman <Tim_at_SageLogix.com>
Date: Sun, 18 Aug 2002 05:58:20 -0800
Message-ID: <F001.004B8645.20020818055820@fatcity.com>


I'm sure that each source is accompanied by some type of explanation? Surely neither would make such a bald sweeping statement without some substantiation?

I would agree with your summarization of Mr. Lewis's position. The type of I/O activity associated with online redo log files (i.e. high volumes of sequential write I/O) suggests an unbuffered type of file. Storing the data written by LGWR in any form of a file-system buffer-cache (which is characteristic of most file-systems) would be a waste of time and processing. Caching is associated with things which you want to write once and then read many, many times. This is not the case with redo data written down by LGWR. At best, it will be read only once more (by the ARCH process), and certainly not in a timely-enough fashion as to justify buffering it...

As the name suggests, "raw" devices (a.k.a "logical volumes") are just that -- an unmanaged, unbuffered, "raw" access to the underlying I/O subsystem. There are also file-systems with which the caching can be turned off (i.e. DEC/Compaq/HP AdvFS, etc), so like any simple sweeping statement, the real answer is more complicated, but in general Mr. Lewis's statements are certainly more correct than the other mystery document.

Please consider the possibility also that you may have misunderstood the message in the "couple years old document by Oracle", because the interpretation you stated is so far from correct...

> Hi,
> please give your opinion.
> I am reading Jonathan Lewis book Practical Oracle 8i.
> He says redo log definitly should be placed on raw devices.
> On the other hand I have a couple of years old document by Oracle
> stating redo logs definitly do not belong on raw devices.
> So who is right?
> Antje Sackwitz
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> --
> Author: Antje Sackwitz
> INET: Stejskal_at_t-online.de
>
> Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> San Diego, California -- Public Internet access / Mailing Lists
> --------------------------------------------------------------------
> 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).

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Tim Gorman
  INET: Tim_at_SageLogix.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
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 Sun Aug 18 2002 - 08:58:20 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US