Hertz are hertz, but don't make the assumption that on intel, adding a
processor doubles
the throughput. The target market for x86 emulation would be older processors
in need of
upgrade - not new ones. Nobody who wants to impress their management should be
recommending this solution until it's
On Wed, Nov 12, 2008 at 12:10 AM, Alan Cox <[EMAIL PROTECTED]> wrote:
> Good JIT engine emulation these days is a lot lot better than that. You
> do need a lot of memory to make that work well.
Some popular modern techniques around JIT do not work very well in a
shared environment, as we found wi
Well, this sure looks like a way of turning the BSOD into the BIG BLUE
Screen Of Death.
On Tue, Nov 11, 2008 at 10:47 AM, jose raul baron <[EMAIL PROTECTED]> wrote:
> Hi List, we have recently been asked to virtualize a windows server under
> SLES9 Linux running in a z/VM environment. In other wor
> Simple arithmetic: Emulation normally takes an order of magnitude more
> cycles. Thus a
> workload that consumes 10% of a 2Ghz intel processor would consume 1/2 of a
> z10 IFL. Even
Good JIT engine emulation these days is a lot lot better than that. You
do need a lot of memory to make that
On Tuesday, 11/11/2008 at 02:36 EST, Berry van Sleeuwen
<[EMAIL PROTECTED]> wrote:
> I too don't consider x86 emulation to meet business needs. It would be
> interesting to see if it will run from a specialist point of view. But
> for production loads I want the software to run with as little over
Eh, Barton, you mean 1 Intel 2ghz core, don't you? Assuming an Intel server to
be 2 quad-core sockets, that would be almost 1 z10 IFL versus 1 Intel server.
Or do I completely misunderstand you?
Best regards,
Pieter Harder
[EMAIL PROTECTED]
tel +31-73-6837133 / +31-6-47272537
-Oorspronkel
Academic in the fact that it wouldn't make sense from a business perspective.
Simple arithmetic: Emulation normally takes an order of magnitude more cycles.
Thus a
workload that consumes 10% of a 2Ghz intel processor would consume 1/2 of a z10
IFL. Even
with the new low price for z10 IFL, ma
On Tue, Nov 11, 2008 at 2:25 PM, Staller, Allan <[EMAIL PROTECTED]> wrote:
> IANA IBM'er. However, I have been looking extensively into
> virtualization as a means to save my company some serious GREEN dollars,
> not funny money and not eco-dollars.
While you can (and will!) save GREEN dollars (I
Hello Florian,
IMO it is the other way around. Microsoft doesn't build it's software
for zseries therefore it will not run on zseries. Perhaps IBM would sell
zseries machines to run windows, just like they do for linux, but then
Microsoft would have to build the software for this CPU. And also oth
IANA IBM'er. However, I have been looking extensively into
virtualization as a means to save my company some serious GREEN dollars,
not funny money and not eco-dollars.
VMware also provides much of the same functionality, but only runs on
*IX based HW(Sun, HP,...). I do not think (and I could be w
On Tue, 11 Nov 2008, Calvin wrote:
> Hello John
>
> I was under the impression that you guys can possibly help with any
> mainframe related questions. If you do not know you could find out to
> someone who might have the answer. I read in this forum when someone was
> trying to IPL guest z/OS with
On Tuesday, 11/11/2008 at 12:49 EST, Florian Bilek
<[EMAIL PROTECTED]> wrote:
> I would really like to understand IBM's view on this issue. Is it from a
> business point of view that Windows or Intel (x86) is not considered as
a
> serious environment for running mission critical applications??
No,
On Nov 11, 2008, at 11:15 AM, David Boyes wrote:
On 11/11/08 11:57 AM, "Alan Altmark" <[EMAIL PROTECTED]> wrote:
I keep
waiting for an operating system written in Java with a
byte-code-interpreting CPU!
Well, there is the p-system OS. Not the IBM P, and not written in
Java, but
very platform
On 11/11/08 11:57 AM, "Alan Altmark" <[EMAIL PROTECTED]> wrote:
> I keep
> waiting for an operating system written in Java with a
> byte-code-interpreting CPU!
Well, there is the p-system OS. Not the IBM P, and not written in Java, but
very platform independent, and there were systems manufactur
Hello Alan,
I would like to come back on your statement:
It's certainly
interesting from an academic perspective, but it doesn't meet the needs of
business.
I would really like to understand IBM's view on this issue. Is it from a
business point of view that Windows or Intel (x86) is not conside
Regarding QEMU, the overhead is not that big. On a z/990 on an LPAR with 2
IFLs it consumed 50% during installation. Afterwards as long as its idle it
doesn't take too much processing power.
But as mentioned in one of the other posts, z/VOS sounds really interesting.
BR Florian
On Tue, Nov 11,
On Tuesday, 11/11/2008 at 11:24 EST, jose raul baron
<[EMAIL PROTECTED]> wrote:
> Certainly I agree the proper thing would be to migrate the workload
instead
> of keeping the same philosophy but... the customer is always right.
(cough)
As many customers who have talked to me can attest, no, they
If you are interested, I ported QEMU on Debian s390. The problem with QEMU
is that it does not compile with gcc 4.x compiler. It uses gcc 3.4.6. I am
running it under z/VM 5.3 on a z/990. I used a Standard Windows 2003 server.
The installation run about 8 hours but finally it worked. The performanc
> So my question is: do any of you know any way to achieve this "invention"
> our customer demands from us ?
Bochs works, but at a tremendous overhead cost -- on a MP3000 it was just
about 2 full CPUs to run one Windows instance. It's not as bad now, but it
still is a pretty significant hit. QEMU
I believe, if memory serves, that they were going to make some sort of official
announcement about their product sometime in November, with GA in the 1st
quarter of 2009.
Note that the Mantissa x86 virtualization engine runs under CMS, so there is no
need (or
expense) to acquire a Linux on zSeri
>>> On 11/11/2008 at 9:45 AM, "Lee, Gary D." <[EMAIL PROTECTED]> wrote:
> Our kernel-s390x is currently at 2.6.5-7.314 (due to patches being
> installed),
> but
> the available kernel-source is 2.6.5-7.97
>
> If there is a package for 2.6.5-7.314-s390x, where do I find it?
> Or will 2.6.5-7.97
Oh, I see. I feared that.
Certainly I agree the proper thing would be to migrate the workload instead
of keeping the same philosophy but... the customer is always right.
OK, thank you very much for your help, gentlemen.
BRGDS,
José R. Barón
Dpto. Sistemas
CALCULO S.A.
Tel. 91 330 86 44
E-m
Not a direct answer, but Mantissa is working on x/86 under z/VM. Target
release date is 1Q2009. Virtualization rate is estimated to be about 700
images per z/9 engine. You can extrapolate for z/890 or z/10.
http://www.mantissa.com/products/UV
Hi List, we have recently been asked to virtualize a
>>> On 11/11/2008 at 10:47 AM, jose raul baron <[EMAIL PROTECTED]> wrote:
-snip-
> So my question is: do any of you know any way to achieve this "invention"
> our customer demands from us ?
Currently, no, there is no way to do this. That may be changing in the future,
if the folks from Mantissa
Hi List, we have recently been asked to virtualize a windows server under
SLES9 Linux running in a z/VM environment. In other words:
- A zSeries system running
- a z/VM system running
- a SLES9 Linux running
- a windows (XP server I suppose)
My question is: How can I virtualize
On Tue, 2008-11-11 at 16:24 +0200, Calvin wrote:
> mainframe related questions [...]
> Which forum can I send my request to?
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu
Our kernel-s390x is currently at 2.6.5-7.314 (due to patches being installed),
but
the available kernel-source is 2.6.5-7.97
If there is a package for 2.6.5-7.314-s390x, where do I find it?
Or will 2.6.5-7.97 work for the lin_tape install?
Thanks for any and all help.
Gary Lee
Senior System Prog
Hello John
I was under the impression that you guys can possibly help with any
mainframe related questions. If you do not know you could find out to
someone who might have the answer. I read in this forum when someone was
trying to IPL guest z/OS without success due to the console config.
Which fo
Hi, Craig.
I'm sorry that I missed your original post. Mike MacIsaac forwarded
some of this thread to me. (And then I signed back up to the list.
Long story!)
The usual thing that happens in Linux (and most Unix) is for the root
filesystem to be mounted read-only when /sbin/init gets control.
David Stuart wrote:
We run CRAM jobs here, now, all the time.
In our case, it stand for Cross Region Access Method (Model 204, anyone?).
Yep. I know of one site, ever.
At the time ADABAS (3.2.1 and 4.1) were just about all over Canberra,
mostly on FACOM hardware (IBM had offended public serva
Gentry, Stephen wrote:
There is some doc. on the internet regarding this technology.
http://archive.computerhistory.org/resources/text/NCR/NCR.CRAM.1960.1026
46240.pdf
(Watch the wrap).
Anybody want to write a driver for this? 8-)
Got some hardware?
--
Cheers
John
-- spambait
[EMAIL PRO
Mark Post wrote:
On 11/10/2008 at 4:43 PM, Bernard Wu <[EMAIL PROTECTED]> wrote:
Hi List,
From a Linux guest, when I send an email and a file as an attachment to MS
Outlook, the attachment is displayed as part of the message body.
How can I send it so that Outlook receives it as an attached fil
Calvin wrote:
Hello
I am running z/OS as a guest and have problem in attaching the OSA devices,
I can only see the channel path with no devices, and how can I configure the
CHPID to be able to see those devices?
We discuss Linux here. z/OS when it relates to communication with Linux.
--
C
> The fsck is not what is causing the I/O errors. When you see "fsck
succeeded" it's done and over with. Let me guess, though.
Well I guess I can rule out the "fsck" then. Why the difference between
SLES9 and SLES10 though?
> Is this an ext3 file system (or reiserfs), and are you mounting it a
Thanks Mark,
>nail -a attachment -s "subject line" [EMAIL PROTECTED] < /dev/null
That did the trick and if you want something in the body of the message :
echo "message in body" | mail -s "subject line" [EMAIL PROTECTED]
And it works without using uuencode .
Thanks to all who replied.
Bernie Wu
Craig,
You should also update the FSTAB in the following way:
/dev/disk/by-path/ccw-0.0.0xxx-part1/ ext2ro,acl,user_xattr 0
0
For me this works fine. The xxx has also an ext2 filesystem on it. I had
also ext3 in the beginning and disabled journaling with tunefs. 0 0 at te
end avoid
36 matches
Mail list logo
