Real life TCO based on real life numbers.

Real life TCO based on real life numbers.

Post by cybea » Sat, 04 Jan 2003 08:55:37



Yeah, MS has reports that they paid for that claim Windows has lower TCO,
but Linux has real life case studies that show Linux TCO is less than 1/2
that of windows...

http://cin.earthweb.com/article.php/10493_1477911

 
 
 

Real life TCO based on real life numbers.

Post by Erik Funkenbusc » Sat, 04 Jan 2003 09:19:31



> Yeah, MS has reports that they paid for that claim Windows has lower
> TCO, but Linux has real life case studies that show Linux TCO is less
> than 1/2 that of windows...

> http://cin.earthweb.com/article.php/10493_1477911

Why do you people keep rehashing the same tired stories over and over again?
Can't you find any new ones?  This particular story was reported here many
times already, and it was largely discredited then because the study failed
to take many costs into account and claimed ridiculous things like Windows
requires 1:10 administrator/machine ratio (I don't know any shops with any
large number of servers that have less than 1:25 ratio, some as high as
1:1000.)

 
 
 

Real life TCO based on real life numbers.

Post by Shane Phelp » Sat, 04 Jan 2003 11:53:31




>>Yeah, MS has reports that they paid for that claim Windows has lower
>>TCO, but Linux has real life case studies that show Linux TCO is less
>>than 1/2 that of windows...

>>http://www.veryComputer.com/

> Why do you people keep rehashing the same tired stories over and over again?
> Can't you find any new ones?  This particular story was reported here many
> times already, and it was largely discredited then because the study failed
> to take many costs into account and claimed ridiculous things like Windows
> requires 1:10 administrator/machine ratio (I don't know any shops with any
> large number of servers that have less than 1:25 ratio, some as high as
> 1:1000.)

I'd agree with Erik on this one. The summary is just that, a summary,
and the actual report seems mysteriously difficult to dig up.

An additional complicating factor is that the report was funded by IBM,
who just happens to be pushing Linux * the server side (especially
web servers) in competition with MS and Sun Microsystems. It probably
contains a certain amount of payback to Sun for one of Sun's earlier TCO
reports where Sun's boxes came out much cheaper than IBM's ;-)

There are lies, damned lies and TCO "studies". Statistics without the
raw data don't count, and even with the raw data can be misleading (eg
how critical are the systems' response times and downtime?)

 
 
 

Real life TCO based on real life numbers.

Post by cybea » Sat, 04 Jan 2003 12:29:02




>> Yeah, MS has reports that they paid for that claim Windows has lower
>> TCO, but Linux has real life case studies that show Linux TCO is less
>> than 1/2 that of windows...

>> http://cin.earthweb.com/article.php/10493_1477911

> Why do you people keep rehashing the same tired stories over and over
> again?

Because you guys keep rehashing the tired old MS paid for reports.

Quote:> Can't you find any new ones?  

Can't you find anything that actually supports your view, other than your
own personal views?

Quote:> is particular story was reported here
> many times already, and it was largely discredited then because the
> study failed to take many costs into account and claimed ridiculous
> things like Windows requires 1:10 administrator/machine ratio (I don't
> know any shops with any large number of servers that have less than 1:25
> ratio, some as high as 1:1000.)

It was not discredited. There was no supporting evidence that discredited
it, only personal claims from people that have already shown they would
say just about anything to bash Linux. That hardly discredits the
article.
 
 
 

Real life TCO based on real life numbers.

Post by cybea » Sat, 04 Jan 2003 12:34:29



> An additional complicating factor is that the report was funded by IBM,

So? The winvocates seem to think it OK to drag up MS paid for studies.
 
 
 

Real life TCO based on real life numbers.

Post by Mr. Berserk » Sat, 04 Jan 2003 17:54:44




> > Yeah, MS has reports that they paid for that claim Windows has lower
> > TCO, but Linux has real life case studies that show Linux TCO is less
> > than 1/2 that of windows...

> > http://cin.earthweb.com/article.php/10493_1477911

> Why do you people keep rehashing the same tired stories over and over again?
> Can't you find any new ones?  This particular story was reported here many
> times already, and it was largely discredited then because the study failed
> to take many costs into account and claimed ridiculous things like Windows
> requires 1:10 administrator/machine ratio (I don't know any shops with any
> large number of servers that have less than 1:25 ratio, some as high as
> 1:1000.)

BS. Go back to Mordor, do not pass Lothlorien, do not collect 200
letters of maggot folk scrawling.

...But feel free to collect 200 clues, even though it would do you
little good.

 
 
 

Real life TCO based on real life numbers.

Post by Daero » Sat, 04 Jan 2003 22:56:39






>> > Yeah, MS has reports that they paid for that claim Windows has lower
>> > TCO, but Linux has real life case studies that show Linux TCO is less
>> > than 1/2 that of windows...

>> > http://cin.earthweb.com/article.php/10493_1477911

>> Why do you people keep rehashing the same tired stories over and over again?
>> Can't you find any new ones?  This particular story was reported here many
>> times already, and it was largely discredited then because the study failed
>> to take many costs into account and claimed ridiculous things like Windows
>> requires 1:10 administrator/machine ratio (I don't know any shops with any
>> large number of servers that have less than 1:25 ratio, some as high as
>> 1:1000.)

>BS. Go back to Mordor, do not pass Lothlorien, do not collect 200
>letters of maggot folk scrawling.

>...But feel free to collect 200 clues, even though it would do you
>little good.

Why do you Wintrolls persist in coming around here spreading the usual
fud ?  Is this report recent enough for you Eric ?

www.vnunet.com/News/1137722  -   16-12-2002

"..IDC findings questioned by, well, pretty much everyone really .."

"..analysts, resellers and developers have all questioned the findings
of IDC's Microsoft-sponsored paper which concluded that Windows is
better value than Linux over a five-year period .."

"..  while Windows .. potentially raising staffing costs.. "
                 Andy Butler, vice president at  Gartner

[ The following statement from Al Gillen of IDC is curious as it seems
to contradict itself. ]

 "It wasn't that one Linux person costs 30 per cent more but that
costs could be as much as 30 per cent higher for the staffing element
.."

".. It might take three people to support a Linux server, but two for
Windows. .."

Might it ? This is total bullshit. A more accurate statement is ONE
Linux staff per FIVE Windows staffers.

Some more quotes -

". Everybody knows that you have to go on paying money for upgrades
unless you switch away from Microsoft ."
                                Simon Welch  -  SUN

"..engineer certification is just another way of making money for
Microsoft.."               Mike Lawrence  -  Hewlett Packard

( keep saying TCO.. keep saying TCO.. keep saying TCO.. keep saying
TCO.. keep saying TCO.. )

 
 
 

Real life TCO based on real life numbers.

Post by Shane Phelp » Sun, 05 Jan 2003 00:11:42




>>An additional complicating factor is that the report was funded by IBM,

> So? The winvocates seem to think it OK to drag up MS paid for studies.

Any sponsored report will be biased, so it's not right to accept one
vendor's bias and reject another's.

I explained it in my post. This was partly a go at Microsoft, and partly
payback to Sun for an earlier Sun TCO report showing that a particular
Sun solution was cheaper than IBM's. Sun's TCO report was actually one
of the more credible (less incredible?) in that it focussed on a very
narrow problem domain and showed raw figures as well as summary statistics.

They must have gone to great lengths to get the Sun costs so high. Sun's
hardware is typically twice the cost of comparable commodity Intel
servers, and running costs are similar for any *n?x system. All I can
think of is that the Sun sites used Oracle, commercial web / Java
servers and older commercial firewalls. It's very easy to skew the
figures with a couple of $50k firewalls against a pair of OpenBSD PCs.

TCO is an important factor, but by no means the only one. Some firms
value reliability and continuity of service highly enough to willingly
pay the massive additional costs required to get each of those extra 9s
after the decimal point. 99.9% is easy, 99.99% is getting expensive,
99.999% is enormously expensive, and most wouldn't even *think* of 99.9999%
TCO without reliability figures is just a number!!!!

 
 
 

Real life TCO based on real life numbers.

Post by cybea » Sun, 05 Jan 2003 01:33:41



> Any sponsored report will be biased, so it's not right to accept one
> vendor's bias and reject another's.

Exactly my point. If the wintrolls belive the MS sponsored reports, then,
for them, sponsored reports are OK and the should accept this report.
Nuff said.
 
 
 

Real life TCO based on real life numbers.

Post by T.C. » Sun, 05 Jan 2003 05:00:51




>> Yeah, MS has reports that they paid for that claim Windows has lower
>> TCO, but Linux has real life case studies that show Linux TCO is less
>> than 1/2 that of windows...

>> http://cin.earthweb.com/article.php/10493_1477911

> Why do you people keep rehashing the same tired stories over and over
> again? Can't you find any new ones?  This particular story was reported
> here many times already, and it was largely discredited then because the
> study failed to take many costs into account and claimed ridiculous things
> like Windows requires 1:10 administrator/machine ratio (I don't know any
> shops with any large number of servers that have less than 1:25 ratio,
> some as high as 1:1000.)

Why are you still here on C.O.L.A. then?

If your so confident Linux is finished and our stories are old and dead,
they *WHY* hang around?

Take up masterbation!

Do something meaningful with your life man!

T.C.

 
 
 

Real life TCO based on real life numbers.

Post by T.C. » Sun, 05 Jan 2003 05:01:02





>>>Yeah, MS has reports that they paid for that claim Windows has lower
>>>TCO, but Linux has real life case studies that show Linux TCO is less
>>>than 1/2 that of windows...

>>>http://www.veryComputer.com/

>> Why do you people keep rehashing the same tired stories over and over
>> again? Can't you find any new ones?  This particular story was reported
>> here many times already, and it was largely discredited then because the
>> study failed to take many costs into account and claimed ridiculous
>> things like Windows requires 1:10 administrator/machine ratio (I don't
>> know any shops with any large number of servers that have less than 1:25
>> ratio, some as high as 1:1000.)

> I'd agree with Erik on this one. The summary is just that, a summary, and
> the actual report seems mysteriously difficult to dig up.

> An additional complicating factor is that the report was funded by IBM,
> who just happens to be pushing Linux * the server side (especially
> web servers) in competition with MS and Sun Microsystems. It probably
> contains a certain amount of payback to Sun for one of Sun's earlier TCO
> reports where Sun's boxes came out much cheaper than IBM's ;-)

> There are lies, damned lies and TCO "studies". Statistics without the raw
> data don't count, and even with the raw data can be misleading (eg how
> critical are the systems' response times and downtime?)

You completely crazy.

T.C.

 
 
 

Real life TCO based on real life numbers.

Post by Shane Phelp » Sun, 05 Jan 2003 06:15:45






>>>>Yeah, MS has reports that they paid for that claim Windows has lower
>>>>TCO, but Linux has real life case studies that show Linux TCO is less
>>>>than 1/2 that of windows...

>>>>http://www.veryComputer.com/

>>>Why do you people keep rehashing the same tired stories over and over
>>>again? Can't you find any new ones?  This particular story was reported
>>>here many times already, and it was largely discredited then because the
>>>study failed to take many costs into account and claimed ridiculous
>>>things like Windows requires 1:10 administrator/machine ratio (I don't
>>>know any shops with any large number of servers that have less than 1:25
>>>ratio, some as high as 1:1000.)

>>I'd agree with Erik on this one. The summary is just that, a summary, and
>>the actual report seems mysteriously difficult to dig up.

>>An additional complicating factor is that the report was funded by IBM,
>>who just happens to be pushing Linux * the server side (especially
>>web servers) in competition with MS and Sun Microsystems. It probably
>>contains a certain amount of payback to Sun for one of Sun's earlier TCO
>>reports where Sun's boxes came out much cheaper than IBM's ;-)

>>There are lies, damned lies and TCO "studies". Statistics without the raw
>>data don't count, and even with the raw data can be misleading (eg how
>>critical are the systems' response times and downtime?)

> You completely crazy.

> T.C.

Thank you. I'll take that as a compliment.
 
 
 

Real life TCO based on real life numbers.

Post by SINNE » Tue, 07 Jan 2003 17:57:02


While strolling through comp.os.linux.advocacy, cybear was overheard
plotting:

Quote:> Yeah, MS has reports that they paid for that claim Windows has lower TCO,
> but Linux has real life case studies that show Linux TCO is less than 1/2
> that of windows...

> http://cin.earthweb.com/article.php/10493_1477911

more numbers from January 2003

http://tinyurl.com/44ty

--
David
http://www.mameworld.net/ MameWorld, The largest MAME resource on the net
http://www.mandrakelinux.com/ Activation Free since November 1998
http://www.tombstones.org.uk/burners.php - Free CD MAME ROM Offer

 
 
 

Real life TCO based on real life numbers.

Post by alt » Tue, 07 Jan 2003 23:25:12



> more numbers from January 2003

> http://tinyurl.com/44ty

"The cost of acquisition of software, hardware--all the things you buy up
front--that's a minority element of the total cost of ownership of any
operating system," he said. "Whether you pay $2,000 for a Windows license
or $49.95 for a boxed copy of Linux, over the course of its lifetime, that
ends up being a minor cost."

Of course, what he hasn't mentioned is:

1. "Forced" upgrades every 2 or 3 years by vendors
2. Retraining costs on a new OS for admins and users alike
3. System Failures due to poor quality software from $MONOPOLY

when your whole company isn't working because your server went down, that's
expensive.

 
 
 

Real life TCO based on real life numbers.

Post by SINNE » Wed, 08 Jan 2003 00:33:00


While strolling through comp.os.linux.advocacy, alt was overheard plotting:

Quote:> Of course, what he hasn't mentioned is:

> 1. "Forced" upgrades every 2 or 3 years by vendors
> 2. Retraining costs on a new OS for admins and users alike
> 3. System Failures due to poor quality software from $MONOPOLY

what he HAS mentioned:

According to the study, the three-year cost of a 100,000-hit processing
unit was significantly different among the systems:

Solaris: $561,520
Windows: $190,662
Linux: $74,475

--
David
http://www.mameworld.net/ MameWorld, The largest MAME resource on the net
http://www.mandrakelinux.com/ Activation Free since November 1998
http://www.tombstones.org.uk/burners.php - Free CD MAME ROM Offer

 
 
 

1. A Story From Real Life

Here is an encouraging one.

I've been asked to help a colleague who is deploying a web app at one of
our locations that is neither here nor there, meaning it ain't his site and
it ain't mine.

This probably means a platform change, as I like to have a development
platform that mimics as much as possible the deployment platform.  Well,
not liking to be in the dark, I raised a fuss that we had to test this
remote place out as quick as possible, to determine how responsive they
are, and so forth, to make sure no deadlines would slip in the bureaucratic
morass.

They send me the network diagram and I'm looking through the servers
available to us.  Solaris, solaris, solaris, solaris, linux, linux, etc.  
By long force of habit, I'm looking for the NT/2k box (this is for
$EMPLOYER after all).  Because of course, that would let us prototype at
home on the exact deployment platform (I'm a real cynic on cross-platform
tools, many of them work 99% admirably, but that 1% will kill you
sometimes).

But no NT. No Win2k.  My mind actually slipped at this point and I did not
know what we might do.  Since we are pure MS, could we still get a mirror
platform for development that looked like the deployment platform?

Well, duh.  We'll take the linux choice, naturally, and now it's time to
set up my first official linux box at $EMPLOYER.

One box at a time.
--
Ken
Linux, the more you learn, the more you love

2. downloads slow with 2.2 kernel

3. A real-life experience

4. POSSIBLE VIRUS ?

5. Aw or DOOL-Soap love =Real life love

6. tty error on console + openwin problem

7. Practical (real-life) Unix Security

8. MINI LINUX install help!

9. Any LINUX installations in real life environments?

10. using 2 TB in real life

11. Linux is real life

12. ISA Server in real life

13. Linux is real life.