---Original message---
Newsgroups: bit.listserv.ibm-main
From: timothy.sipp...@us.ibm.com (Timothy Sipples)
Date: 5 Oct 2011 23:59:45 -0700
Local: Thurs, Oct 6 2011 8:59 am
Subject: Re: ZDNET actually says something nice about IBM LINUX
I have to respond to one point Roger Bowle
How many Virtual Linux Servers per IFLHm with the usual answer it
"it depends". Today I am running one z9BC IFL and have up 55+ Virtual Linux
Servers. That is, 3 LPARS (BUILD, TEST, & Prod). As to what they do is more
relevant than the quantity. Have 2 production DB2 LUW server
Mainframe economics continue to improve, and it's important to take
advantage of that if you can, when you can.
With respect to the z10 BC and its IFLs, let's consider a
back-of-the-envelope exercise for WebSphere Application Server workloads.
I'm going to use IBM's LSPR PCI metric as a proxy for
I have to respond to one point Roger Bowler made (again).
Roger, it's impossible to configure a z114 at capacity setting A00 without
either one IFL or one ICF -- and you probably know better. IBM's starting
mainframe configuration is either a single IFL model or a capacity setting
A01 model. Yes,
Yes, sure you can run thousands of guests on one IFL. But running what?
I really can't go into specifics but lets say that we are running less than
20 servers per z10 BC IFL(a mix of production and test). These are mainly
WAS and a some Oracles.
As for the question asked about 30% less of what...
On Wed, 5 Oct 2011 15:42:16 +0200, Vernooij, CP - SPLXM
wrote:
>30% of what you estimated or 30% lower than you estimated?
>Can you give figures? I understood one could run 'thousands' of guests.
Here is what IBM marketing have been claiming:
"IBM calculates that workloads from as many as 300
30% of what you estimated or 30% lower than you estimated?
Can you give figures? I understood one could run 'thousands' of guests.
Kees.
"Mike Shorkend" wrote in message
news:...
> A very rough estimate of about 30%.
>
> On Wed, Oct 5, 2011 at 10:00 AM, Vernooij, CP - SPLXM
> wrote:
>
> > "M
Mike Shorkend wrote:
> Our current issue is that the ratio of zLinux guests per IFL is much
> lower than what was expected.
Vernooij, CP - SPLXM wrote:
> Out of curiosity: how low is 'much lower'?
Mike Shorkend wrote:
>A very rough estimate of about 30%.
Mike,
IBM marketing materials claim t
A very rough estimate of about 30%.
On Wed, Oct 5, 2011 at 10:00 AM, Vernooij, CP - SPLXM wrote:
> "Mike Shorkend" wrote in message
> news:<2726868143899905.wa.mikeshorkend@bama.ua.edu>...
> >
> > At the time it was definitely the right move. For those who have not
> seen my presentation -
"Mike Shorkend" wrote in message
news:<2726868143899905.wa.mikeshorkend@bama.ua.edu>...
>
> At the time it was definitely the right move. For those who have not
seen my presentation - at the time we had no other option but to use
zLinux as we had no room at our DR site for any more blade type
At the time it was definitely the right move. For those who have not seen my
presentation - at the time we had no other option but to use zLinux as we had
no room at our DR site for any more blade type servers.
Our current issue is that the ratio of zLinux guests per IFL is much lower than
what
>>> On 10/4/2011 at 03:19 AM, Sheldon Davis wrote:
> Hi
>
> Does anyone have a method of comparing x86 Linux vs IFL Linux? (Performance
> and cost)
Interesting question coming from someone at Isracard. Mike Shorkend has given
a number of presentations at SHARE about Isracard's experiences ru
There have been several published studies showing that using a
zero-based comparison, the z/Linux solution is cheaper.
Check the archives. At least one person has stated this formally and
offered his analysis to others.
Unfortunately, I have misplaced my copy and can't tell you where the
"break ev
On Mon, 3 Oct 2011 13:25:19 -0600, Mark Post wrote:
>As always, makes me wonder what else they got wrong.
Like this:
IBM measures use of its mainframe systems in millions of instructions per
second (MIPS).
--
Tom Marchant
-
In <9744671434806595.wa.sdavisisracard.co...@alabamamaps.ua.edu>, on
10/04/2011
at 02:19 AM, Sheldon Davis said:
>Does anyone have a method of comparing x86 Linux vs IFL Linux?
>(Performance and cost)
First define your workload; the results will vary depending on whether
you are CPU bound or
I believe the problem is not with Linux kernel itself, but what you run
under...
ITschak
On Tue, Oct 4, 2011 at 9:19 AM, Sheldon Davis wrote:
> Hi
>
> Does anyone have a method of comparing x86 Linux vs IFL Linux? (Performance
> and cost)
>
> Thanks
>
> Sheldon Davis
>
> ---
Hi
Does anyone have a method of comparing x86 Linux vs IFL Linux? (Performance and
cost)
Thanks
Sheldon Davis
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the messa
It's Linux, not LINUX. And it's not IBM Linux, it's Linux for IBM's System z.
IBM doesn't have their own Linux distribution.
It would have been nice if Kusnetzky got IBM's branding right. As always, makes
me wonder what else they got wrong.
Mark Post
Linux on Mainframes - an IBM update
By Dan Kusnetzky | October 3, 2011, 3:03am PDT
Summary: Many Linux proponents appear to equate Linux with the use of industry
standard X86 systems. IBM demonstrated that its Mainframes are a growing
portion of the market.
IBM presented an update on Linux on its
19 matches
Mail list logo
