Bruno Sugliani writes:
>I do not believe anyone here would run production on a single
>engine box or run Dev and Test on that single engine machine.

Almost everyone here would. These are mainframes, with PR/SM and LPARs
proven to Common Criteria EAL5+ certification standards. Regardless of the
operating system(s) running in particular LPARs.

The prototypical Linux on z implementation consists of three z/VM LPARs:
two for production, one for dev/test.

Clark Morris writes:
>Why would running on system z require fewer guests than running in a
>virtual machine environment on another platform?

I'm going to agree and disagree with Mark Post on this one. I'll agree to
the extent that the number of guests (virtual images) is often similar.
However, there are at least a couple reasons why there would be fewer
guests on a zEnterprise server:

1. If you were running clustering to protect primarily against hardware
failures, then a move to zEnterprise is likely to reduce guests. Production
and production test guests would be reduced by half or more in that case.

2. If you had a correlation between virtual images and virtualization (or
server) instances -- for management servers of various kinds, for example
-- then a move to zEnterprise is likely to reduce the number of those
management-related guests.

3. If you had a server sprawl problem (virtual or otherwise), it's quite
likely that reducing the number of physical servers all by itself will help
enforce some greater management discipline and reduce the number of guests.
(You might call this the "spring cleaning" effect.) One reason is just the
physical challenges in cycling hundreds (or more) versus one or a few
physical servers. With greater numbers of physical servers you're
inevitably going to have many servers in the racks that are provisioned but
not dispatched/online. It's just basic fleet management logic. Remember,
there's a delay before an image (or server) actually goes into active
service and another delay before it's retired. You can't have absolutely
100% perfect alignment -- you always have to over-provision to some extent
-- but you can get a lot closer to 100% efficiency with a smaller number of
bigger servers that virtualize better across projects and domains.

There are some other reasons, but those are three examples.

--------------------------------------------------------------------------------------------------------
Timothy Sipples
Consulting Enterprise IT Architect (Based in Singapore)
E-Mail: [email protected]
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to