RE: Filesystem blocksize for archive logs on AIX

From: Ric Van Dyke <>
Date: Thu, 24 Oct 2013 12:35:07 -0500
Message-ID: <>

It's likely not something to get too worked up over. With today's intelligent frames and massive frame caching, the block size of the disks isn't nearly the concern it used to be. And frankly some of the "proof" of these blocks size recommendations of the past are a bit shaky. Unless you are going to totally by pass the frame and go back to JBOD, the frame is going to be doing most of the really IO and Oracle will for the most part be reading and writing to the frame cache.

About a thousand years ago when I was working on V6 in VMS, everything was in 512. Oracle does have its roots in VMS, so as you say it's likely that at the base 512 would likely be a good setting. But today most IOs are done in about a 2k size so 512 is likely a bit small.

-----Original Message-----

[] On Behalf Of Rich Jesse Sent: Thursday, October 24, 2013 1:02 PM To:
Subject: Filesystem blocksize for archive logs on AIX

Hey all,

I'm configuring new DBs on a new AIX 7.1 box. Instead of cloning our current setup on AIX 5.3, I'm revisiting all options for both flexibility and performance.

The docs
( and MOS articles like 1507249.1 are great, but one item I can't find anywhere is a recommendation for a filesystem blocksize for the archive logs. 4K for datafiles, 512 for online redos and controlfile, but nothing for archives.

512 would seem to be the likely answer, but it's been too many years since my v7.3 DBA classes and I can't remember how the archiver does its IO (sorry, Joe!). The docs don't seem to have that answer, either.

I'm also open to any other advice for AIX-specific setup that's not referenced in the above docs (e.g. JFS2 noatime/rbrw for $ORACLE_HOME?). If it makes a difference, we'll be using XIV for storage. Seriously looking forward to playing with that puppy... :D



-- Received on Thu Oct 24 2013 - 19:35:07 CEST

Original text of this message