-Original Message-
From: The IBM z/VM Operating System
[mailto:ib...@listserv.uark.edu] On Behalf Of Richard Troth
Sent: 05 January 2010 01:04
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: Moving on: The University of Maine (System)
shuts down mainframe
Actually, Jack nailed it.
Hi Ed!
Can you share this exec here with us at the NYC PD computer center?.
Thank you!
Luis Ticona
646-610-5304
-Original Message-
From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On
Behalf Of Edward M Martin
Sent: Tuesday, July 14, 2009 11:20 AM
To:
Hi Ed. If you are willing to share I would like a copy of your exec as well
please. Thanks.
david.jua...@va.gov
David Juárez
Department of Veterans Affairs
z/OS and z/VM Systems Programmer
512-326-6116 Work
-Original Message-
From: The IBM z/VM Operating System
Hello All,
We are upgrading our z9BC to a z10BC and increasing the IFLs from two to three.
I am planning on modifying the Linux guest definitions to utilize the third
IFL. Are there any performance issues or problems adding a third IFL? I can't
imagine this would do anything but help, I'm
TIA,
The z10BC IFLs are about 50% faster that the z9BC IFL.
So, you are already getting more CPU power.
If you over assign IFLs to a guest, it may just cause spin locks preventing
other guests from getting CPs that the first guests has been assigned but is
not using.
See if they really need the
I wouldn't add the 3rd IFL to the Directory for the Linux guests. Just
leave it as another engine for CP to dispatch work on.
On Tue, Jan 5, 2010 at 10:39 AM, Stock, Roger W rst...@bu.edu wrote:
TIA,
The z10BC IFLs are about 50% faster that the z9BC IFL.
So, you are already getting more CPU
Nick,
Disclaimer: I'm not a performance geek.
I presume that you believe that you need to add a third virtual processor
to each Linux guest's virtual machine definition (directory entry) for it
to use the added third real processor. That's not true. Unless
constrained in some way, CP will
Hi Mike and thanks for the update.
According to our performance product we do NOT need an additional IFL at this
time but it was included as part of the z10 upgrade so I wanted to utilize it.
Thanks,
Nick
-Original Message-
From: The IBM z/VM Operating System
On Tuesday, 01/05/2010 at 12:25 EST, Harris, Nick J.
nhar...@txfb-ins.com wrote:
According to our performance product we do NOT need an additional IFL at
this
time but it was included as part of the z10 upgrade so I wanted to
utilize it.
It will be utilized even though you do not define
As others have said - you don't need to change any guests for z/VM to
utilize the additional IFL - it will do so all on it's own.
Scott
On Tue, Jan 5, 2010 at 10:24 AM, Harris, Nick J. nhar...@txfb-ins.comwrote:
Hi Mike and thanks for the update.
According to our performance product we do
Won¹t adding an additional IFL to the z/VM system increase the license
charge for your linux distribution?
On 1/5/2010 at 12:35 PM, Neale Ferguson ne...@sinenomine.net wrote:
Won*t adding an additional IFL to the z/VM system increase the license
charge for your linux distribution?
Yes, unless it's a non-commercial distribution.
Mark Post
To say it with other words: CP is not really dispatching userids, but it
dispatches virtual CPUs, and only virtual CPU's that want to work (CP knows
when a virtual CPU places itself in wait, so it won't dispatch sleeping
CPUs).
An example: suppose you have 3 Linuxes, each with 2 virtual CPUs (that
Side note..
We are adding 2 CP engines to our z10 next week. I *believe* we can vary
these engines onto VM without an IPL.
Someone very knowledgeable told me he was 99% sure that would work. :-)
Anyone out there know differently?
(We currently are using 6 of the existing engines on the
How is the amount of CP dump space calculated?
We have a system that seems to be using at lot more than the others.
Everything is pretty much identical across the LPARs HW wise (access to all the
same devices, etc) and SW wise (z/VM 5.4 RSU 0902).
Here's an example.
System A:
q dump
Dave --
This is ridiculous. Do you really want to start a religious war on
the value of Windows ... on this discussion list?
To your two points, cost and flexibility, I dig in for further
analysis as follows.
COST
Windows is not more cost effective. It carries a finer grained price
point
There was a problem with spxtape which I thought was fixed by z/VM 5.4.
But as we see it, on z/VM 5.2
SPXTAPE tried to buffer as much as possible.
As spxtape is a CP function, all of its buffers end up requiring slots on the
CP Dump dataset.
When spxtape finishes, there is no automatic way
But isn't the spxtape use cleared up by set dump off#set dump dasd ?
That just made it grow...
Q RECORDING shows nothing pending retrieval.
I also cleaned up spool so both systems have roughly the same number of spool
files.
Marcy
This message may contain confidential and/or privileged
Well, best of luck.
Of course it’s disheartening to see another VM shop bite the dust but that’s
the way things seem to be going these days.
Perhaps this is a sign of the times.
Of course there will, for the most part, be mainframes
around for those organizations which require lots
Look in the CP Planning and Administration in the Allocating DASD Chapter
under Spooling Space and allocating space for CP hard Abends. There is a
table for estimating space allocation, but it is all dynamic depending on
the number of users and the amount of storage allocated.
Regards,
Jihad K.
Well, I think I found my own answer.
VM64477 - HIPER and not yet on an RSU.
I'll confirm with a PMR to IBM.
Marcy
This message may contain confidential and/or privileged information. If you
are not the addressee or authorized to receive this for the addressee, you must
not use, copy,
I also cleaned up spool so both systems have roughly the same number of
spool files.
In the future, roughly the same number of spool files isn't he same as
knowing the size of the spool files.
From a Class D userid, 'CP Query RDR EXP' (and also for PUN, and PRT) will
help provide the answer
Richard,
I meant to say the perceived cost was less, but its the perceived cost
thats important.
As you suggest the granularity of costs is also a big factor and is probably
one of the major factors responsible for the dearth of servers (Solaris as
well as Windows) we have in our Data Centre.
Howard,
Opps I forgot to wish Wayne all the best and enjoys his life beyond
mainframes. It is sad to see all the folks leaving the list who helped me
when I was a VM systems programmer starting out all those years ago.
I am not sure how many Mainframes there are left in the North West of
Nick,
z9BC to z10BC : Last time I looked at an upgrade, processor speed will be
better but, check what was ordered. Engine speeds can depend on which model of
z10BC. As others(Roger,Mark,Alan,...) have already said, NO need to change the
directory entries first shot out of the barn, except for
Christy,
IF the LPAR is defined with spare processors (reserved), you will be able to
vary the added processors online. Next though, you mentioned '2 CP engines',
are they IFL's or General Purpose processors? Mixing the two? We only have a z9
and z/VM 5.4. Never tried it.
Doug
- Original
On 1/5/2010 at 7:35 PM, Doug Shupe dsh...@bellsouth.net wrote:
Nick,
z9BC to z10BC : Last time I looked at an upgrade, processor speed will be
better but, check what was ordered. Engine speeds can depend on which model
of z10BC.
Specialty engines, such as the IFL, _always_ run at full
Meant GP not CP :-) Thanks!
From: Doug Shupe dsh...@bellsouth.net
28 matches
Mail list logo