RE: Bare-bones high-availability system

Jan,

Replies inline:

Regards,
Andy Nolen-Parkhouse

> On Behalf Of Jan Megannon
> Subject: Bare-bones high-availability system
> 
> Hi,
> 
> With the announcement of the new systems and the way they are packaged,
> is there a way of configuring the following? Or may it just be
> considered a 'nice-to-have'?
> 
> The Standard Edition comes with zero interactive (but enough to run a
> console)
> The Enterprise Edition, on certain models, comes with dormant
> processors
> for CUoD.

I think you have a basic misunderstanding here.  The models which come
with dormant processors (825, 870, and 890) will do so regardless of
whether they are ordered with the Standard or Enterprise packages.

> It would be quite nifty to be able to purchase a machine for
> high-availability with a product such as *noMAX (or MIMIX, Vision or
> Data Mirror). This would be a Standard Edition machine. The replication
> occurs in batch.

This would certainly work.

> The system will have been configured with sufficient interactive
> processors, which will also be dormant. These will then be activated in
> the case of a disaster, as this will be when you require interactive
> power on the DR system, serving as CUoD processors.

Processors and interactive performance are really two different things
and are not directly related.  You either buy the Enterprise package or
you don't.  You either activate the processors (temporary or permanent)
or you don't.  You do have a good point that you could undersize a DR
machine with only enough processor power to replicate.  Then you could
temporarily activate additional processors for the duration of your
disaster.  But you would not be able to just switch 5250 on and off under
IBM's current offering.

> Is there a place for such an animal? Am I explaining myself correctly?
> As far as I can see, IBM should not lose: they will move some more
> machines and have the maintenance on these in any case.

IBM has, in the past, specifically addressed this issue with discounts or
rebates on DR machines of a minimum configuration purchased with an
approved high-availability product.  I've forgotten most of the details
but I seem to recall that it involved allowing your system to be
monitored (like PM/400) so that it doesn't exceed certain performance
limits except for a few days a year.  Now that they've reconfigured the
way they package the systems, I would look for them to come up with new
incentives for customers to implement high-availability.  As you say, it
is to their advantage and to purchase an entire enterprise suite for use
during only disasters may be financially prohibitive.  The relevant
announcements for these previous offers are 301-106 (July 2001), 302-015
(February 2002), and 302-063 (April 2002).  Because they've done it in
the past, I would expect them to do it again.  For anyone looking to
implement a new high-availability installation, it might be worth waiting
to see what they do, or try to force their hand.

> What do the genii think?
> 
> Cheers.
> 
> Jan.
> 
> _______________________________________________
> This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
> list
> To post a message email: MIDRANGE-L@midrange.com
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo.cgi/midrange-l
> or email: MIDRANGE-L-request@midrange.com
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/midrange-l.