z/OS Architecture Diagram

[Lindy Mayfield]

Currently I’m working with Linux and I found this diagram which I like very much.

http://flylib.com/books/3/475/1/html/2/images/0131777203/graphics/02fig01.gif

I’ve searched google images and more places looking for something similar as that for z/OS. I didn’t find one.

Somewhere I read an opinion that z/OS was the most complex piece of technology mankind has ever come up with. The more I learn, the more I think that is pretty valid.

I think somehow a z/OS diagram in the same style as the Linux kernel diagram would be similar, but even after working with MVS/z/OS my entire life, so much of it is still a mystery. How can something so complex be so stable and efficient?

As the world moves more to Linux, in order to survive I too keep up with and constantly improve my Linux skills as mainframe work declines, which is par for the course, of course, but who knows, RedHat running on system z in some sort of cloud environment may be as fast or faster and/or more efficient than pizza boxes which are basically “throw-away” when something goes wrong, the only difference is the cost of it all, z/VM and the system z hardware.

Any links to pictures or comments between the differences between z/OS and Linux kernel architecture are of course welcome.

Kind regards,
Lindy

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to list...@listserv.ua.edu with the message: INFO IBM-MAIN

[Gerhard Adam]

Here's a similar type of diagram from OS/360

http://www.lbdsoftware.com/os360.jpg

Adam

[Gibney, Dave]

At that very high level, is there a difference :)

> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:IBM-...@LISTSERV.UA.EDU]
> On Behalf Of Lindy Mayfield
> Sent: Thursday, September 01, 2016 5:55 PM
> To: IBM-...@LISTSERV.UA.EDU
> Subject: z/OS Architecture Diagram
>
> Currently I’m working with Linux and I found this diagram which I like very
> much.
>
>
>

> https://urldefense.proofpoint.com/v2/url?u=http-
> 3A__flylib.com_books_3_475_1_html_2_images_0131777203_graphics_02
> fig01.gif&d=DQIGaQ&c=C3yme8gMkxg_ihJNXS06ZyWk4EJm8LdrrvxQb-
> Je7sw&r=u9g8rUevBoyCPAdo5sWE9w&m=2TWgkvRAYvCWitPcdZYItJ6l6XBb
> pSHRjNL1GeyLRQE&s=9-Tm_R8KruU5PuXvDkCUWMzlFw1F6QNmtPB-
> G7uevTY&e=

[Tom Brennan]

Hmm... maybe all we'll find are charts from way way back. Here's one
from my hardcover book "Invitation to MVS Logic and Debugging" by Harry
Katzan Jr. and Davis Tharayil. But that's from 1984!
http://www.mildredbrennan.com/mvs/figure1.2.png

Maybe the ABC's of Programming redbooks have something.

[Lindy Mayfield]

Exactly, Dave! :-) And it was partially my point of asking.

There is something about z/OS (MVS, etc) that makes it special over other operating systems (I believe), and I just can't put my finger on it precisely. I figured a high level diagram like that might show something. (They do a bit, I think, the ones that Gerhard and Tom sent. Thank you both.)

Regards,
Lindy

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-...@LISTSERV.UA.EDU] On Behalf Of Gibney, Dave
Sent: perjantaina 2. syyskuuta 2016 4.28
To: IBM-...@LISTSERV.UA.EDU
Subject: Re: z/OS Architecture Diagram

At that very high level, is there a difference :)

> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:IBM-...@LISTSERV.UA.EDU]
> On Behalf Of Lindy Mayfield
> Sent: Thursday, September 01, 2016 5:55 PM
> To: IBM-...@LISTSERV.UA.EDU
> Subject: z/OS Architecture Diagram
>
> Currently I’m working with Linux and I found this diagram which I like
> very much.
>

[Charles Mills]

http://orig08.deviantart.net/36fb/f/2012/070/8/2/blue_ball_machine_by_andreadionne-d4sg8sa.gif

It's Friday ...

Charles

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-...@LISTSERV.UA.EDU] On Behalf Of Lindy Mayfield
Sent: Friday, September 02, 2016 2:27 PM
To: IBM-...@LISTSERV.UA.EDU
Subject: Re: z/OS Architecture Diagram

Exactly, Dave! :-) And it was partially my point of asking.

There is something about z/OS (MVS, etc) that makes it special over other operating systems (I believe), and I just can't put my finger on it precisely. I figured a high level diagram like that might show something. (They do a bit, I think, the ones that Gerhard and Tom sent. Thank you both.)

[Lindy Mayfield]

not anymore in GMT+2 :-(

but good diagram. I can use that. :-)

/Lindy

[Scott Ford]

Charles/Lindy,

I missed this one yeah man , that was my week...😀

Scott

On Fri, Sep 2, 2016 at 5:45 PM, Lindy Mayfield <Lindy.M...@sas.com>
wrote:

[Joel C. Ewing]

Here's an imperfect digital image of a 1991 IBM wall poster of "MVS/ESA
V4 System Flow" complete with IBM Document number (lower-left):
http://jce-atreides.getmyip.com:8080/rd/Ay3qdFdu/ESA_Sys_Flow.jpg
My recollection is that the original was 3' - 4' on a side.

The multiplicity of isolated user virtual address spaces which gave MVS
its name and strong inter-user protection properties is shown by stacked
"User Area"s and "Extended User Area"s, a diagram subtlety which would
likely be overlooked by someone who didn't understand the significance.

I remember seeing other MVS-OS/360-z/OS wall-poster structure diagrams,
but this seems to be the only one I bothered to preserve.
Joel C. Ewing

On 09/01/2016 08:29 PM, Tom Brennan wrote:
> Hmm... maybe all we'll find are charts from way way back. Here's one
> from my hardcover book "Invitation to MVS Logic and Debugging" by
> Harry Katzan Jr. and Davis Tharayil. But that's from 1984!
> http://www.mildredbrennan.com/mvs/figure1.2.png
>
> Maybe the ABC's of Programming redbooks have something.
>
> Gerhard Adam wrote:
>> Here's a similar type of diagram from OS/360
>>
>> http://www.lbdsoftware.com/os360.jpg
>>
>> Adam
>>


--

Joel C. Ewing, Bentonville, AR jce...@acm.org

[Charles Mills]

That one is pretty good, except for the lack of USS and the fact that it only goes to 2GB.

Charles

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-...@LISTSERV.UA.EDU] On Behalf Of Joel C. Ewing
Sent: Saturday, September 03, 2016 2:50 PM
To: IBM-...@LISTSERV.UA.EDU
Subject: Re: z/OS Architecture Diagram

Here's an imperfect digital image of a 1991 IBM wall poster of "MVS/ESA
V4 System Flow" complete with IBM Document number (lower-left):
http://jce-atreides.getmyip.com:8080/rd/Ay3qdFdu/ESA_Sys_Flow.jpg
My recollection is that the original was 3' - 4' on a side.

The multiplicity of isolated user virtual address spaces which gave MVS its name and strong inter-user protection properties is shown by stacked "User Area"s and "Extended User Area"s, a diagram subtlety which would likely be overlooked by someone who didn't understand the significance.

[Bill Woodger]

Is it just me, or does it look like Nucleus and Expanded are full of Post-it (R) notes?

[Joel C. Ewing]

On 09/03/2016 06:05 PM, Bill Woodger wrote:
> Is it just me, or does it look like Nucleus and Expanded are full of Post-it (R) notes?
>
>

I guessing that's intended to convey pageable page frames in REAL and
EXPANDED physical memory, and that a portion of REAL memory (where there
is an absence of Post-it notes) is occupied by non-pageable Nucleus
code; but yeah those "page frame boundaries" do look a little
imprecisely aligned, which would have greatly complicated algorithms for
Real Memory Management. :)
Joel C. Ewing

--
Joel C. Ewing, Bentonville, AR jce...@acm.org

[Charles Mills]

I thought those were the multiple address spaces. There's only one nucleus, but there's a lot of user region address spaces.

Charles

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-...@LISTSERV.UA.EDU] On Behalf Of Joel C. Ewing
Sent: Saturday, September 03, 2016 4:41 PM
To: IBM-...@LISTSERV.UA.EDU
Subject: Re: z/OS Architecture Diagram

On 09/03/2016 06:05 PM, Bill Woodger wrote:
> Is it just me, or does it look like Nucleus and Expanded are full of Post-it (R) notes?
>
>
I guessing that's intended to convey pageable page frames in REAL and EXPANDED physical memory, and that a portion of REAL memory (where there is an absence of Post-it notes) is occupied by non-pageable Nucleus code; but yeah those "page frame boundaries" do look a little imprecisely aligned, which

[Bill Woodger]

Yes, it's the irregularity which draws me to it. The rest of the diagram I could do with IBM templates plus those terminals/consoles and the printers (I don't have a template for those). Then the little yellow pieces seem to have been done entirely by hand. I was wondering if the lack of regularity had significance beyond "I was really tired by the time I came to finishing it off".

[Joel C. Ewing]

The "multiple address spaces" are depicted in the yellow rectangle top
middle, which distinguishes between those parts of virtual storage that
are common to all address spaces vs the "User" areas of virtual storage
that have multiple occurrences. The rectangle bottom center with no
color fill and yellow "Post-its" appears to represent the physical
hardware (CPs, memory, channels), and since address spaces are a logical
concept not physical and no address space could exist only in EXTENDED
storage, I think a physical page frame is the more likely intended
interpretation for the Post-its. The dashed line inside "REAL" implies
the "NUCLEUS" occupies a sub part of REAL memory. Although I've always
just thought of the Nucleus from an Address Space view as a set of
page-fixed virtual pages, in retrospect since it must be available very
early it makes sense that it should be set up as contiguous pages in
real memory
Joel C. Ewing

On 09/03/2016 06:52 PM, Charles Mills wrote:
> I thought those were the multiple address spaces. There's only one nucleus, but there's a lot of user region address spaces.
>
> Charles
>
> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:IBM-...@LISTSERV.UA.EDU] On Behalf Of Joel C. Ewing
> Sent: Saturday, September 03, 2016 4:41 PM
> To: IBM-...@LISTSERV.UA.EDU
> Subject: Re: z/OS Architecture Diagram
>
> On 09/03/2016 06:05 PM, Bill Woodger wrote:
>> Is it just me, or does it look like Nucleus and Expanded are full of Post-it (R) notes?
>>
>>
> I guessing that's intended to convey pageable page frames in REAL and EXPANDED physical memory, and that a portion of REAL memory (where there is an absence of Post-it notes) is occupied by non-pageable Nucleus code; but yeah those "page frame boundaries" do look a little imprecisely aligned, which
>
>

--
Joel C. Ewing, Bentonville, AR jce...@acm.org

[Charles Mills]

Remember when the Data Areas manual had an overview diagram in the front that showed how the various control blocks pointed to each other? It's been a while. I have the OS/390 V1R2 Data Areas (1996) and it has no such diagram.

Charles

[Anthony L. Zak]

The 'FE Handbook - System / 360 Operating System' circa early 70's had the control block layouts of the major control blocks and charts of the relationships, plus service aid doc.  All of this was in 367 pages in an approximately 3" X 6" binder.
/s/ Anthony L. Zak

[Dana Mitchell]

From Greg Price's web site:

I remember receiving a copy of this one as part of an IBM MVS class in the early 80's

http://www.prycroft6.com.au/misc/download/GV25-6719-0_MVSflow_repaint.gif

Here's a relatively up to date version

http://www.prycroft6.com.au/misc/download/zos_system_overview_poster2.pdf

Dana