On 24 Feb 2008 11:37:54 -0800, [EMAIL PROTECTED] (Gerhard Adam)
wrote:
I agree, however I'm surprised that no one has mentioned the obvious point
that needs to be considered. Chargeback shouldn't be based on usage, but
rather on the capacity that has been reserved for the anticipated load.
How to determine what the charge will be is both a political and an
accounting issue.
With politics being the major one.
Resource-based chargeback is only understandable by IT, and not in every
case.
I agree, however I'm surprised that no one has mentioned the obvious point
that needs to be
-
Too busy driving to stop for gas!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at
: Price of CPU seconds
-
Too busy driving to stop for gas!
--
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
I have argued for many years that charge-back should *not* include CPU seconds
or any other resource measure. Charge back by service. $.50 per transaction.
$.20 per policy. $500 per bed per month.
I have been making the same argument since 1981.
Users should be charged on something
Yes , it is the question of the communication with Linux and NT people.
I wanted to explain to my collegues , why the 0.5 % constant CPU usage for
an idle server is a matter in a large z/OS system.
An argument would be, if I could say: in a week it is nearly an hour,
and an hour CPU in a
In
[EMAIL PROTECTED],
on 02/21/2008
at 09:20 PM, Ted MacNEIL [EMAIL PROTECTED] said:
Regardless of what you think of me, keep your abusive ad hominum attacks
to yourself.
PKB.
--
Shmuel (Seymour J.) Metz, SysProg and JOAT
ISO position; see
In [EMAIL PROTECTED], on 02/19/2008
at 03:53 PM, Miklos Szigetvari [EMAIL PROTECTED]
said:
If someone can tell me the price of a CPU second in a larger z/OS system
(we discuss currently if 0.5% CPU usage is relevant or not )
There is no one size fits all. You amortize the cost of whatever
In
[EMAIL PROTECTED],
on 02/20/2008
at 06:41 PM, Ted MacNEIL [EMAIL PROTECTED] said:
Voodoo mathematics.
No, just a lack of reading comprehension. Which part of If all other
things are equal didn't you understand?
--
Shmuel (Seymour J.) Metz, SysProg and JOAT
ISO position; see
Which part of If all other things are equal didn't you understand?
The part about the fact that all other things are never equal, and the missing
assumption that was later stated.
Which part of reading an entire thread did you not understand.
Regardless of what you think of me, keep your
Date:Wed, 20 Feb 2008 13:53:14 -0500
From:Thompson, Steve [EMAIL PROTECTED]
Subject: Re: Price of CPU seconds
I have argued for many years that charge-back should *not* include CPU
seconds or any other resource measure. Charge back by service. $.50 per
transaction. $.20 per policy
I think if you're simply trying to educate someone familiar with
distributed servers, I would ask something like this:
Are servers free?
And (hopefully) they'll answer, No. Next question:
So how would you measure their costs? If it's a multi-user server, as most
servers are, how would you
Hi
Tom Marchant wrote:
On Wed, 20 Feb 2008 11:27:35 -0800, Gerhard Adam wrote:
First of all these comments weren't intended as a personal attack, and I
never indicated that YOU were ridiculous, but rather than the proposed
mechanism for costing CPU seconds was.
The primary reason they are
I wanted to explain to my collegues , why the 0.5 % constant CPU usage for an
idle server is a matter in a large z/OS system.
I honestly don't believe that it is a big deal.
Measurement and forecasting tools are not that granualar, and accuracy (at
best) is only plus or minus 5-10%.
So, if
-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Gerhard Adam
Sent: Tuesday, February 19, 2008 5:16 PM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: Price of CPU seconds
That example stated that $800 per CPU/hr was the cost for a machine
(undefined
On Tue, 19 Feb 2008 15:15:41 -0800, Gerhard Adam wrote:
So, the more processors you have to split the costs across, the lower
the CPU/hr charge may be. And those charges are based on the SMF
collected times (since that is what fed the accounting system).
It is clearly ridiculous since the
If all other things are equal, the cost per CPU second for a machine with two
processors is half the cost per CPU second of a machine with one processor.
Voodoo mathematics.
If I start a second processor, I have to at least pay for the hardware upgrade
and increased s/w costs.
So, I don't
-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Ted MacNEIL
Sent: Wednesday, February 20, 2008 12:42 PM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: Price of CPU seconds
If all other things are equal, the cost per CPU second for a machine
with two
Since what I have said is so ridiculous, why don't you take a crack at
answering the original question? Then we can all take pot shots at what
you say, pointing out that the example you used, based on a real system,
is absolutely _ [fill in the blank].
First of all these comments weren't
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of Ted MacNEIL
If all other things are equal, the cost per CPU second for a
machine with two processors is half the cost per CPU second
of a machine with one processor.
Voodoo mathematics.
Is that anything like
On Wed, 20 Feb 2008 18:41:52 +, Ted MacNEIL [EMAIL PROTECTED]
wrote:
If all other things are equal, the cost per CPU second for a machine with
two processors is half the cost per CPU second of a machine with one
processor.
Voodoo mathematics.
Anything beyond our comprehension is
OK, all you guys are right. There is no way to do charge-back
accounting. All the formulae are wrong regardless of what they are.
First of all no one said you can't do chargeback, but only that the
simplistic solutions being proposed aren't accurate.
If you have a formula, I'd love to see it
On Wed, 20 Feb 2008 11:27:35 -0800, Gerhard Adam wrote:
First of all these comments weren't intended as a personal attack, and I
never indicated that YOU were ridiculous, but rather than the proposed
mechanism for costing CPU seconds was.
The primary reason they are ridiculous is that the
If your single-engine machine used, say, 100 MSUs, how many MSUs does the
now twin-engine machine use for the same workload?
Roughly the same.
Assuming NO capacity-priced software, how much is the increase in software
cost?
I'm not sure what your term capacity priced means, but if it means
OK, all you guys are right. There is no way to do charge-back accounting. All
the formulae are wrong regardless of what they are. All these circus bureaus
have been defrauding their clients, all the publicly held companies that are
doing charge-back are somehow cooking their books.
Now, I have
So, I don't understand/agree with your statement.
Compare a z9 BC model Y01 (1 CP) to a model S02 (2 CPs). Both are about 420
MIPS. Can you buy them both for the same price? I don't know.
Okay. Now I understand.
There was a missing assumption, that the single engine processor and the two
Miklos Szigetvari wrote:
Hi
If someone can tell me the price of a CPU second in a larger z/OS system
(we discuss currently if 0.5% CPU usage is relevant or not )
100$.
Justification: none. vbg
But seriously: it strongly depends. You cannot say any reasonable answer
without providing many
Hi
If someone can tell me the price of a CPU second in a larger z/OS system
(we discuss currently if 0.5% CPU usage is relevant or not )
--
Miklos Szigetvari
Development Team
ISIS Information Systems Gmbh
tel: (+43) 2236 27551 570
Fax: (+43) 2236 21081
E-mail: [EMAIL PROTECTED]
Info:
If someone can tell me the price of a CPU second in a larger z/OS system
Unfortunately, that is a loaded question.
You have to have a basis on which to answer that question.
1. Do you count just CPU
2. Do you add in software costs
3. Do you add in staffing costs (and which ones)
4. What other
Well, actually you can purchase a fraction of a CPU. The z9 BC, z890, and z800
all have many settings for their CPU speed. But, on the larger processors you
can't do that. You can't add one CPU at half the speed of the others - they
all have to be the same speed.
Eric
R.S. [EMAIL
-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Miklos Szigetvari
Sent: Tuesday, February 19, 2008 7:30 AM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: [IBM-MAIN] Price of CPU seconds
Hi
For me this more or less clear.
I have here a number
On Tue, 19 Feb 2008 16:30:03 +0100, Miklos Szigetvari wrote:
For me this more or less clear.
I have here a number of collegues from NT and Unix , and they don't
understand why the 0.5% CPU time is a matter:
I see.
I suppose they also don't have any idea how MVS can run all day long at
100%
-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Miklos Szigetvari
Sent: Tuesday, February 19, 2008 8:53 AM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Price of CPU seconds
Hi
If someone can tell me the price of a CPU second in a larger z/OS system
(we
Hi
For me this more or less clear.
I have here a number of collegues from NT and Unix , and they don't
understand why the 0.5% CPU time is a matter:
/Would somebody knowledgeable please explain to me why some host people
get their panties in a knot (I love colorful expressions!) over a
On Tue, 19 Feb 2008 16:30:03 +0100, Miklos Szigetvari wrote:
For me this more or less clear.
I have here a number of collegues from NT and Unix , and they don't
understand why the 0.5% CPU time is a matter:
/Would somebody knowledgeable please explain to me why some host people
get their
The following message is a courtesy copy of an article
that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well.
[EMAIL PROTECTED] (Miklos Szigetvari) writes:
If someone can tell me the price of a CPU second in a larger z/OS system
(we discuss currently if 0.5% CPU usage
-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Miklos Szigetvari
Sent: Tuesday, February 19, 2008 9:30 AM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: Price of CPU seconds
Hi
For me this more or less clear.
I have here a number of collegues
Last year we found ourselves CPU constrained at peak hour and undertook a
series of initiatives to save it.
The metric we used to decide whether a particular initiative was worth taking -
which did not pretend to be other than a very rough guide - was 1p (0.01
pounds UK) per CPU second.
I don't even know the price of CPU firsts.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at
You're not going to have that overhead for only one day. 0.5% of a $10
Million computer is $50,000. Of course, that ignores software costs and
the
other things that have been mentioned. To think only of CPU seconds
trivializes it.
Unfortunately these kinds of calculations are not only
On Tue, 19 Feb 2008 10:49:40 -0800, Gerhard Adam wrote:
You're not going to have that overhead for only one day. 0.5% of a $10
Million computer is $50,000.
... This simple example is based on the notion of the computer costing
$10,000,000 every day.
No, it isn't. It's based upon the total
-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Gerhard Adam
Sent: Tuesday, February 19, 2008 12:50 PM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: Price of CPU seconds
SNIP
One of the other examples uses $0.22/second which results in nearly
You're not going to have that overhead for only one day. 0.5% of a $10
Million computer is $50,000.
... This simple example is based on the notion of the computer costing
$10,000,000 every day.
No, it isn't. It's based upon the total life cost and assumes that the
processor
is kept until
That example stated that $800 per CPU/hr was the cost for a machine
(undefined as to number of CPUs, MSUs, etc.). It also did not state what
the system software cost, etc. It was a number used to give an example.
And it was about what the cost was for an Amdahl 5990-1400 w/ 2 Gig
C-Store and 2
Probably due to the fact that the NT Unix guys get another box when they ask
for one without much questioning. You're dealing with with single
application/single server mentality.
I did a tuning exercise of a Domino Change Management application (non mail
server application) running on
Steve Thompson writes:
Price of CPU is determined by cost of processor (the whole box), all the
software that runs on it, all the power to run the system (CEC, RAID,
ATL, etc.), and then costs of personnel to run the system.
With you so far. That's called total price, assuming you've got all the
46 matches
Mail list logo