Well, for what it's worth --

I have a 3584, 16 LTO drives (10 LTO4, 6 LTO2), and two SAN switches.

I have all the odd-numbered drives in switch 1 and all the even numbered drives 
in switch 2.

My TSM server has 8 fibers for tape access - 4 in each switch. I've set up 4 
zones in each switch - one per TSM server fiber - with all 8 drives assigned in 
each zone.

And this does give 64 tape devices at the AIX level. The 16 base devices   
(-PRI) are the rmtx devices used in the TSM define path construct.

In peak periods, all 8 paths get beat up pretty heavily, but not so much that 
more paths look like they'd help any.

(I'm using MRTG to capture the traffic statistics by port on the switches and 
then tying the switch ports back to the devices attached at the other end).

Tom Kauffman
NIBCO, Inc

-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of 
Allen S. Rout
Sent: Friday, March 07, 2008 12:39 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: Tape drive zones for FC drives - best practices

Dredging up a conversation from last year.  I know in theory the horse
is dead, but I think I might have seen a twitch.

>> On Thu, 8 Feb 2007 10:35:56 +0100, Remco Post <r.post AT SARA DOT NL> said:

> Really I've never seen anybody recommend anything above one zone per
> HBA.


The 'Implementing IBM Tape in Unix Systems' redbook talks about
dynamic load balancing (per tape open) across all available HBAs.
This leads me to think that, if I have 4 HBAs devoted to tape traffic,
I want to have each one zoned with all my (dual ported, natch) tape
connections.  This sounds sane, so far.  But then I do math and get


8 drives * 2 ports * 4 HBAs = 64 rmt devices for my 8 drives.  I know
that instAtape will collapse those down so I only need to use the -PRI
devices for e.g. paths.  but it just makes me feel odd to have so
many.  So, I'm wondering:

Am I being silly?  Just go ahead and do it?  Should I be zoning some
of the HBAs for port 0 on the tapes, others for port 1?

I have an inclination, which I am working to stifle, to have this neat
spirograph of a venn diagram so that HBA 1 is connected to odd ports
of even tapes, and HBA 2 is connected to ...



- Allen S. Rout
- Maybe that was just maggots?

CONFIDENTIALITY NOTICE:  This email and any attachments are for the
exclusive and confidential use of the intended recipient.  If you are not
the intended recipient, please do not read, distribute or take action in
reliance upon this message. If you have received this in error, please
notify us immediately by return email and promptly delete this message
and its attachments from your computer system. We do not waive
attorney-client or work product privilege by the transmission of this
message.