I think that z/OS and its immediate predecessors already make most of
what is needed available.
Only in special circumstances do I see or, if I see them, read
Shmuel's posts; but I found one---It is one of his characteristic
Nonsense! posts---in this thread. It was directed at an earlier
post of
jwgli...@gmail.com (John Gilmore) writes:
That said, Timothy Sipples is clearly right in general: The use of
significant system-software resources to identify and share
bit-equivalent pages, even big ones, dynamically is not likely to pay
for itself in a z/OS environment. The major
In
cae1xxdhjs_8hfezdeb9la9-ppazqpniusxkdkjpyvcf0yyo...@mail.gmail.com,
on 07/30/2014
at 11:11 AM, John Gilmore jwgli...@gmail.com said:
My comment that elicited his Nonsense! response suggested very
briefly that the device of making and marking a load module or
program object reentrant or,
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of Shmuel Metz
In
cae1xxdhjs_8hfezdeb9la9-ppazqpniusxkdkjpyvcf0yyo...@mail.gmail.com,
on 07/30/2014
at 11:11 AM, John Gilmore jwgli...@gmail.com said:
My comment that elicited his Nonsense! response suggested
Anne Lynn Wheeler wrote:
begin extract
It wasn't [until, jg] well into the MVS release cycle ... that it
dawned on them that they were selecting non-changed, shared, high-use
linkpack pages for replacement before selecting lower use, application
private data pages . . .
/end extract
and it I
On Mon, 28 Jul 2014 08:42:34 -0400, Shmuel Metz (Seymour J.) wrote:
on 07/27/2014 at 01:52 PM, Tim Zielke said:
This is a simplistic approach for sure,
Actually, it's a complicated approach. The simple approach would be to
add support for using shared virtual memory when fetching REFR and
RENT
As a threshold question, in the real world how many 4K, 1MB, and 2GB frames
(for example) are and remain bit-identical duplicates of each other in a
typical operating machine? What percentage of total memory does that
duplication represent? Do IBM's processor cache algorithms already
de-duplicate?
john.archie.mck...@gmail.com (John McKown) writes:
Sounds a bit like a z/VM DCSS.
re:
http://www.garlic.com/~lynn/2014i.html#85 z/OS physical memory usage with
multiple copies of same load module at different virtual addresses
an issue with DCSS (and each virtual address space not being able
On Sun, 27 Jul 2014 22:01:02 -0400, Jim Mulder wrote:
There is some sharing across address spaces via z/OS Unix of program
objects that live in the file system, using what I think of as shared
private.
Peter Relson
z/OS Core Technology Design
Is this only as a consequence of fork(), or
000433f07816-dmarc-requ...@listserv.ua.edu (Paul Gilmartin) writes:
Understood. When I first explored the S/370 instruction set, it
appeared to me that the S/360 designers had anticipated support
for location-independent code (as envisioned by Lynn Wheeler),
but software never realized
On Mon, 28 Jul 2014 09:07:16 -0500, Paul Gilmartin paulgboul...@aim.com wrote:
On Sun, 27 Jul 2014 22:01:02 -0400, Jim Mulder wrote:
There is some sharing across address spaces via z/OS Unix of program
objects that live in the file system, using what I think of as shared
private.
Peter
In 3699753162619660.wa.paulgboulderaim@listserv.ua.edu, on
07/25/2014
at 03:48 PM, Paul Gilmartin
000433f07816-dmarc-requ...@listserv.ua.edu said:
As Joe D'Alessandro pointed out, with a citation in the PoP, there's
considerable hardware support for this function.
You don't need
In
829975020fe05747b35aa575a61a6ac114642...@dm1prd6402mb047.021d.mgd.msft.net,
on 07/27/2014
at 01:52 PM, Tim Zielke tim.zie...@aon.com said:
This is a simplistic approach for sure,
Actually, it's a complicated approach. The simple approach would be to
add support for using shared virtual
This is a simplistic approach for sure, but it seems to me if z/OS supported
something like this, there would be a significant amount of physical memory
savings on the system. At least at our shop.
1. (Assuming z/OS does not already do this) Set up infrastructure, so every
PDS/PDSE member
tim.zie...@aon.com (Tim Zielke) writes:
4. (Nth time load of a load module into the system by secondary
address space B) When loading the module into the virtual address
space B and physical memory, make sure any virtual pages for the load
module that do not need to be altered for B and were
There is some sharing across address spaces via z/OS Unix of program
objects that live in the file system, using what I think of as shared
private.
Peter Relson
z/OS Core Technology Design
--
For IBM-MAIN subscribe / signoff /
There is some sharing across address spaces via z/OS Unix of program
objects that live in the file system, using what I think of as shared
private.
Peter Relson
z/OS Core Technology Design
For that type of sharing, a shared program appears
at the same virtual address in all of the
On Sun, Jul 27, 2014 at 9:01 PM, Jim Mulder d10j...@us.ibm.com wrote:
There is some sharing across address spaces via z/OS Unix of program
objects that live in the file system, using what I think of as shared
private.
Peter Relson
z/OS Core Technology Design
For that type of sharing, a
john.archie.mck...@gmail.com (John McKown) writes:
Sounds a bit like a z/VM DCSS.
but that isn't how i originally implemented it, first on cp67/cms and
then moved over to vm370/cms ... as part of paged mapped filesystem for
cms
http//www.garlic.com/~lynn/submain.html#mmap
but as i've
In 1406226811.17805.251.camel@localhost, on 07/24/2014
at 02:33 PM, David Andrews d...@lists.duda.com said:
I was told once that one of the advantages of HASP was that it
essentially reintroduced partitions to MVT (after MVT got rid of
partitions). Was this really so?
Partitions are MFT; MVT
In 1406227187.17805.254.camel@localhost, on 07/24/2014
at 02:39 PM, David Andrews d...@lists.duda.com said:
On Thu, 2014-07-24 at 13:27 -0400, Anne Lynn Wheeler wrote:
MFT-II/RASP work was ignored
That tickles a neuron. Was there not a RASP component to OS/VS1
JES?
No. Could you be
The technology is available on another IBM platform, Power7 with PowerVM
Enterprise edition has a feature called Active Memory Deduplication. I have no
idea how widespread the adoption and use of this feature is.
Active Memory™ Deduplication is a feature that is available on systems that
have
On Thu, 2014-07-24 at 19:30 -0400, Shmuel Metz (Seymour J.) wrote:
In 1406227187.17805.254.camel@localhost, on 07/24/2014
at 02:39 PM, David Andrews d...@lists.duda.com said:
Was there not a RASP component to OS/VS1 JES?
No. Could you be thinking of RES?
Could be. I read a lot of code in
Thank you all for the responses. I appreciate it! My question was specific to
load modules in a PDS/PDSE on z/OS, so I believe no one has contradicted Jim's
answer is no below for that use case.
Thanks,
Tim
-Original Message-
From: IBM Mainframe Discussion List
Avoidance would seem to be better than detection and 'deduplication'.
If 1) the execution loader has brought a load module or program object
into storage and 2) that executable is marked refreshable and/or
reentrant, the execution loader will not bring second or subseq
On 7/25/14, David Andrews
sipp...@sg.ibm.com (Timothy Sipples) writes:
z/VM performs such magic in at least three different ways: Discontiguous
Saved Segments, Named Saved Systems, and VM Data Spaces. These mechanisms
are probably somewhat relevant to z/OS when it operates as a z/VM guest.
I hate to disagree with Jim
On Fri, 25 Jul 2014 14:57:15 -0500, Tom Marchant wrote:
On Fri, 25 Jul 2014 15:17:48 -0400, Shmuel Metz (Seymour J.) wrote:
on at 09:42 AM, John Gilmore said:
If 1) the execution loader has brought a load module or program
object into storage and 2) that executable is marked refreshable
I do not believe there is any such functionality.
snip
I had a question about how z/OS handles physical memory for the same load
module that is loaded at different virtual addresses in different virtual
address spaces.
For example, let's say on the same z/OS system I have the following:
CICS
On Thu, 24 Jul 2014 15:54:28 +, Tim Zielke wrote:
I had a question about how z/OS handles physical memory for the same load
module that is loaded at different virtual addresses in different virtual
address spaces.
For example, let's say on the same z/OS system I have the following:
CICS
On Thu, 24 Jul 2014 16:47:54 +, Blaicher, Christopher Y. wrote:
Only way I know of to get different address spaces to use the same physical
memory for a program is to put the module in LPA.
fork(). But in that case, the virtual addresses are identical in different
address spaces. (CoW?
While this is not quite the same thing as 2 CICS regions using a page of the
same load module, take a look in Principles of Operation which talks about how
fork is handled. From page 3-15 of August 2010 version of POP:
The suppression-on-protection function is useful
in performing the POSIX
re:
http://www.garlic.com/~lynn/2014i.html#66 z/OS physical memory usage with
multiple copies of same load module at different virtual addresses
a little more topic drift. part of the MVT storage allocation issues was
with contiguous storage allocation.
quite a few customers were convinced to
On Thu, 2014-07-24 at 13:52 -0400, Anne Lynn Wheeler wrote:
to get around a lot of MVT storage fragmentation
I was told once that one of the advantages of HASP was that it
essentially reintroduced partitions to MVT (after MVT got rid of
partitions). Was this really so? I don't remember enough
Jim Mulder z/OS System Test IBM Corp. Poughkeepsie, NY
I had a question about how z/OS handles physical memory for the same
load module that is loaded at different virtual addresses in
different virtual address spaces.
For example, let's say on the same z/OS system I have the
d...@lists.duda.com (David Andrews) writes:
That tickles a neuron. Was there not a RASP component to OS/VS1
JES?
re:
http://www.garlic.com/~lynn/2014i.html#66 z/OS physical memory usage with
multiple copies of same load module at different virtual addresses
z/VM performs such magic in at least three different ways: Discontiguous
Saved Segments, Named Saved Systems, and VM Data Spaces. These mechanisms
are probably somewhat relevant to z/OS when it operates as a z/VM guest.
I hate to disagree with Jim Mulder. :-) But I'm going to disagree with his
36 matches
Mail list logo
