The references I cited in my previous post are documentation elaborations.
Nothing more, nothing less. IBM did not (generally) recommend multiple JES3
globals in a single Sysplex (which can consist of a single LPAR or not),
past tense, and IBM does not (generally) recommend multiple JES3 globals in
such an environment, present tense. Here's the previous documentation
language, even before elaboration:

"Although it is possible to have multiple JES3 complexes within a Sysplex,
IBM recommends a one-to-one relationship between the JES3 complex and the
Sysplex...."

Here's the revised language:

"IBM recommends a one-to-one relationship between the JES3 complex and the
Sysplex for ease of operations. If you do need to set up multiple global
systems within the same Sysplex, the following considerations apply...."

Though I prefer the revised language, the recommendation hasn't changed. A
recommendation is not a requirement and never was, and a recommendation is
certainly not a technology limitation. One ought to assign meaning and
significance to every word, at least as a starting point. Especially when
the recommendation was preceded by the words "it is possible."

I'm not sure why the previous language (apparently, by some) was misread
and misinterpreted. It certainly shouldn't have been. The previous language
was perfectly clear. Nonetheless, IBM elaborated.

This not exactly new recommendation (not requirement) is cause for
celebration, one would think.

Note that IBM recommended (and recommends) to clients they not follow this
general JES3 recommendation when specifying the requirements for certain
service offerings, as an example.

Let's take a brief look at this "not exactly new" history. I can fairly
easily trace JES3 back a quarter century. (Perhaps somebody else would like
to go back into the pre-Sysplex JES3 era, from 1973 to 1990, to see what
IBM recommended and/or required.)

Sysplex debuted in 1990 with MVS/ESA Version 4. As IBM announcement letter
290-487 helpfully explains, JES3 took advantage of Sysplex from the start.
I don't have convenient access to the MVS/ESA Version 4 documentation at
the moment, but nearly 20 years ago IBM published this redbook, in the
MVS/ESA Version 5 era:

http://www.redbooks.ibm.com/redbooks/pdfs/sg244582.pdf

Section 8.12 reiterates exactly the same JES3 recommendation...and then
proceeds to provide instructions on how to violate IBM's
recommendation. :-) There's even a diagram of that multiple global (and
local) scenario, Figure 55.

A quarter century is a long time to keep holding onto a ~41 year old IT
grudge. :-)

--------------------------------------------------------------------------------------------------------
Timothy Sipples
IT Architect Executive, Industry Solutions, IBM z Systems, AP/GCG/MEA
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