Linux vs. NT performance / Mindcraft results

Linux vs. NT performance / Mindcraft results

Post by A. Steinhof » Wed, 26 May 1999 04:00:00



Bruce Weiner wrote in his rebuttal:
 "Setting the Record Straight: Where ABCnews.com Got It Right and Wrong":

            "2.Mindcraft used a server with a MegaRAID controller
               with a beta driver (which was the latest version
               available at the time of the test) while the PC Week
               server used an eXtremeRAID controller with a fully
               released driver. The MegaRAID driver was single
               threaded while the eXtremeRAID driver was
               multi-threaded."

That means that these Mindcraft guys were aware that they have tested
the LINUX performance with a BETA DRIVER for the MegaRAID controller and
they were aware that this beta driver was just a SINGLE-THREADED driver!!

It's evident that the multi-threaded NT driver for the MegaRAID controller
has a much better performance under havy load as a single-threaded
driver ... so their so called test results are very dubiuos and rigged.

As Lincoln said:"One can cheat the poeple .. but only for a limited time"

Armin

 
 
 

Linux vs. NT performance / Mindcraft results

Post by ?àDé? á??ò » Wed, 26 May 1999 04:00:00


1. In comp.arch.embedded it isn't so interesting...
2.Yes. Linux has some problems with threads...
3.Yes. Linux hasn't problems with drivers for newest devices...(no drivers -
no problems).

    At7

P.S. Don't worry. Linux's enough good OS w/o any tests...


Quote:

>Bruce Weiner wrote in his rebuttal:
> "Setting the Record Straight: Where ABCnews.com Got It Right and Wrong":

>            "2.Mindcraft used a server with a MegaRAID controller
>               with a beta driver (which was the latest version
>               available at the time of the test) while the PC Week
>               server used an eXtremeRAID controller with a fully
>               released driver. The MegaRAID driver was single
>               threaded while the eXtremeRAID driver was
>               multi-threaded."

>That means that these Mindcraft guys were aware that they have tested
>the LINUX performance with a BETA DRIVER for the MegaRAID controller and
>they were aware that this beta driver was just a SINGLE-THREADED driver!!

>It's evident that the multi-threaded NT driver for the MegaRAID controller
>has a much better performance under havy load as a single-threaded
>driver ... so their so called test results are very dubiuos and rigged.

>As Lincoln said:"One can cheat the poeple .. but only for a limited time"

>Armin


 
 
 

Linux vs. NT performance / Mindcraft results

Post by BobX » Wed, 26 May 1999 04:00:00


Linux is only a TOY, and NT is a total JOKE.  You guys should check out
Solaris.

if you guys are driving hardware not supported under Solaris, the
hardware is too new (ergo not tested), or is bad.

period.

BobX


> 1. In comp.arch.embedded it isn't so interesting...
> 2.Yes. Linux has some problems with threads...
> 3.Yes. Linux hasn't problems with drivers for newest devices...(no drivers -
> no problems).

>     At7

> P.S. Don't worry. Linux's enough good OS w/o any tests...


> >Bruce Weiner wrote in his rebuttal:
> > "Setting the Record Straight: Where ABCnews.com Got It Right and Wrong":

> >            "2.Mindcraft used a server with a MegaRAID controller
> >               with a beta driver (which was the latest version
> >               available at the time of the test) while the PC Week
> >               server used an eXtremeRAID controller with a fully
> >               released driver. The MegaRAID driver was single
> >               threaded while the eXtremeRAID driver was
> >               multi-threaded."

> >That means that these Mindcraft guys were aware that they have tested
> >the LINUX performance with a BETA DRIVER for the MegaRAID controller and
> >they were aware that this beta driver was just a SINGLE-THREADED driver!!

> >It's evident that the multi-threaded NT driver for the MegaRAID controller
> >has a much better performance under havy load as a single-threaded
> >driver ... so their so called test results are very dubiuos and rigged.

> >As Lincoln said:"One can cheat the poeple .. but only for a limited time"

> >Armin

 
 
 

Linux vs. NT performance / Mindcraft results

Post by jsnyde » Wed, 26 May 1999 04:00:00


Talk about BRAIN DEAD.  Out of the box Solaris will barely drive a VGA
screen, let alone having any useful drivers for other stuff.


> Linux is only a TOY, and NT is a total JOKE.  You guys should check out
> Solaris.

> if you guys are driving hardware not supported under Solaris, the
> hardware is too new (ergo not tested), or is bad.

> period.

> BobX


> > 1. In comp.arch.embedded it isn't so interesting...
> > 2.Yes. Linux has some problems with threads...
> > 3.Yes. Linux hasn't problems with drivers for newest devices...(no
drivers -
> > no problems).

> >     At7

> > P.S. Don't worry. Linux's enough good OS w/o any tests...


> > >Bruce Weiner wrote in his rebuttal:
> > > "Setting the Record Straight: Where ABCnews.com Got It Right and
Wrong":

> > >            "2.Mindcraft used a server with a MegaRAID controller
> > >               with a beta driver (which was the latest version
> > >               available at the time of the test) while the PC Week
> > >               server used an eXtremeRAID controller with a fully
> > >               released driver. The MegaRAID driver was single
> > >               threaded while the eXtremeRAID driver was
> > >               multi-threaded."

> > >That means that these Mindcraft guys were aware that they have tested
> > >the LINUX performance with a BETA DRIVER for the MegaRAID controller
and
> > >they were aware that this beta driver was just a SINGLE-THREADED
driver!!

> > >It's evident that the multi-threaded NT driver for the MegaRAID
controller
> > >has a much better performance under havy load as a single-threaded
> > >driver ... so their so called test results are very dubiuos and rigged.

> > >As Lincoln said:"One can cheat the poeple .. but only for a limited
time"

> > >Armin

 
 
 

Linux vs. NT performance / Mindcraft results

Post by Armin Steinhof » Wed, 26 May 1999 04:00:00



Quote:

>1. In comp.arch.embedded it isn't so interesting...

Nothing heard about 'embedded NT' ?  ... embedded LINUX ??

Quote:>2.Yes. Linux has some problems with threads...

No ... there are multi-threaded driver of RAID systems for LINUX.

Why has Mindcraft choosen a single-threaded beta driver for the LINUX benchmark?

Was it choosen by chance ??  I don't think so ...

Quote:>3.Yes. Linux hasn't problems with drivers for newest devices...(no drivers -
>no problems).

 ... there is some changing :-)

Armin

 
 
 

Linux vs. NT performance / Mindcraft results

Post by Mav » Wed, 26 May 1999 04:00:00


This bozo is just trying to start a flame war....

Kill this thread now and maybe he'll go play by himself somewhere.

Mav


>Talk about BRAIN DEAD.  Out of the box Solaris will barely drive a VGA
>screen, let alone having any useful drivers for other stuff.



>> Linux is only a TOY, and NT is a total JOKE.  You guys should check out
>> Solaris.

>> if you guys are driving hardware not supported under Solaris, the
>> hardware is too new (ergo not tested), or is bad.

>> period.

>> BobX


>> > 1. In comp.arch.embedded it isn't so interesting...
>> > 2.Yes. Linux has some problems with threads...
>> > 3.Yes. Linux hasn't problems with drivers for newest devices...(no
>drivers -
>> > no problems).

>> >     At7

>> > P.S. Don't worry. Linux's enough good OS w/o any tests...


>> > >Bruce Weiner wrote in his rebuttal:
>> > > "Setting the Record Straight: Where ABCnews.com Got It Right and
>Wrong":

>> > >            "2.Mindcraft used a server with a MegaRAID controller
>> > >               with a beta driver (which was the latest version
>> > >               available at the time of the test) while the PC Week
>> > >               server used an eXtremeRAID controller with a fully
>> > >               released driver. The MegaRAID driver was single
>> > >               threaded while the eXtremeRAID driver was
>> > >               multi-threaded."

>> > >That means that these Mindcraft guys were aware that they have tested
>> > >the LINUX performance with a BETA DRIVER for the MegaRAID controller
>and
>> > >they were aware that this beta driver was just a SINGLE-THREADED
>driver!!

>> > >It's evident that the multi-threaded NT driver for the MegaRAID
>controller
>> > >has a much better performance under havy load as a single-threaded
>> > >driver ... so their so called test results are very dubiuos and
rigged.

>> > >As Lincoln said:"One can cheat the poeple .. but only for a limited
>time"

>> > >Armin

 
 
 

Linux vs. NT performance / Mindcraft results

Post by Dennis Cla » Wed, 26 May 1999 04:00:00


: Linux is only a TOY, and NT is a total JOKE.  You guys should check out
: Solaris.
: if you guys are driving hardware not supported under Solaris, the
: hardware is too new (ergo not tested), or is bad.

  Arrr!  I thinks I smells a TROLL!

DLC
--
------------------------------------------------------------
| Dennis Clark                                             |
| Be well, Do good work, stay in touch -- Garrison Keillor |
------------------------- CUT HERE -------------------------

 
 
 

Linux vs. NT performance / Mindcraft results

Post by Armin Steinhof » Wed, 26 May 1999 04:00:00



Quote:

>This bozo is just trying to start a flame war....

>Kill this thread now and maybe he'll go play by himself somewhere.

This tread doesn't start a 'flame war' .... it shows just facts, if you like
these facts or not.

It raised up just the question about the real performance of the NT server.
Must the NT server be compared with a simple LINUX machine which runs with a
_single-threaded_ RAID driver to show some performance differences ???

Just a dumb question :-)

Armin


>>Talk about BRAIN DEAD.  Out of the box Solaris will barely drive a VGA
>>screen, let alone having any useful drivers for other stuff.



>>> Linux is only a TOY, and NT is a total JOKE.  You guys should check out
>>> Solaris.

>>> if you guys are driving hardware not supported under Solaris, the
>>> hardware is too new (ergo not tested), or is bad.

>>> period.

>>> BobX


>>> > 1. In comp.arch.embedded it isn't so interesting...
>>> > 2.Yes. Linux has some problems with threads...
>>> > 3.Yes. Linux hasn't problems with drivers for newest devices...(no
>>drivers -
>>> > no problems).

>>> >     At7

>>> > P.S. Don't worry. Linux's enough good OS w/o any tests...


>>> > >Bruce Weiner wrote in his rebuttal:
>>> > > "Setting the Record Straight: Where ABCnews.com Got It Right and
>>Wrong":

>>> > >            "2.Mindcraft used a server with a MegaRAID controller
>>> > >               with a beta driver (which was the latest version
>>> > >               available at the time of the test) while the PC Week
>>> > >               server used an eXtremeRAID controller with a fully
>>> > >               released driver. The MegaRAID driver was single
>>> > >               threaded while the eXtremeRAID driver was
>>> > >               multi-threaded."

>>> > >That means that these Mindcraft guys were aware that they have tested
>>> > >the LINUX performance with a BETA DRIVER for the MegaRAID controller
>>and
>>> > >they were aware that this beta driver was just a SINGLE-THREADED
>>driver!!

>>> > >It's evident that the multi-threaded NT driver for the MegaRAID
>>controller
>>> > >has a much better performance under havy load as a single-threaded
>>> > >driver ... so their so called test results are very dubiuos and
>rigged.

>>> > >As Lincoln said:"One can cheat the poeple .. but only for a limited
>>time"

>>> > >Armin

 
 
 

Linux vs. NT performance / Mindcraft results

Post by Scott A. spam-me-some-moo » Wed, 26 May 1999 04:00:00



Quote:

>Bruce Weiner wrote in his rebuttal:
> "Setting the Record Straight: Where ABCnews.com Got It Right and Wrong":

>            "2.Mindcraft used a server with a MegaRAID controller
>               with a beta driver (which was the latest version
>               available at the time of the test) while the PC Week
>               server used an eXtremeRAID controller with a fully
>               released driver. The MegaRAID driver was single
>               threaded while the eXtremeRAID driver was
>               multi-threaded."

>That means that these Mindcraft guys were aware that they have tested
>the LINUX performance with a BETA DRIVER for the MegaRAID controller and
>they were aware that this beta driver was just a SINGLE-THREADED driver!!

>It's evident that the multi-threaded NT driver for the MegaRAID controller
>has a much better performance under havy load as a single-threaded
>driver ... so their so called test results are very dubiuos and rigged.

>As Lincoln said:"One can cheat the poeple .. but only for a limited time"

>Armin

The test was paid for by Microsoft. A more interesting statistic would
be the number of tests that came out favorably for the people who paid
for the test results. Unfortunately, no one appears to want to pay for
that study.

                                              [sam]

 
 
 

Linux vs. NT performance / Mindcraft results

Post by Judah Diamen » Wed, 26 May 1999 04:00:00


Sweeping statements like that usually are by definition wrong. Each OS
has its plusses and minuses. It happens to be that on average, NT, like
all MS products, is unreliable, and I cringe at the thought of
safety-critical apps running on NT (how do you verify an app on an OS
which is TOTALLY closed?!?). Solaris is a very nice OS, used in many
serious RT apps (FAA collision avoidance, for example), but is a royal
pain to configure out of the box. Linux is more user freindly than
Solaris, and while on the one hand doesn't have the same long track
record as Solaris, it is improving much faster, is 100% verifiable
(since you have the source code), and has a slightly larger development
team with quite a bit of tallent, which promises good things to come :-)

                -Judah


> Linux is only a TOY, and NT is a total JOKE.  You guys should check out
> Solaris.

> if you guys are driving hardware not supported under Solaris, the
> hardware is too new (ergo not tested), or is bad.

> period.

> BobX


> > 1. In comp.arch.embedded it isn't so interesting...
> > 2.Yes. Linux has some problems with threads...
> > 3.Yes. Linux hasn't problems with drivers for newest devices...(no drivers -
> > no problems).

> >     At7

> > P.S. Don't worry. Linux's enough good OS w/o any tests...


> > >Bruce Weiner wrote in his rebuttal:
> > > "Setting the Record Straight: Where ABCnews.com Got It Right and Wrong":

> > >            "2.Mindcraft used a server with a MegaRAID controller
> > >               with a beta driver (which was the latest version
> > >               available at the time of the test) while the PC Week
> > >               server used an eXtremeRAID controller with a fully
> > >               released driver. The MegaRAID driver was single
> > >               threaded while the eXtremeRAID driver was
> > >               multi-threaded."

> > >That means that these Mindcraft guys were aware that they have tested
> > >the LINUX performance with a BETA DRIVER for the MegaRAID controller and
> > >they were aware that this beta driver was just a SINGLE-THREADED driver!!

> > >It's evident that the multi-threaded NT driver for the MegaRAID controller
> > >has a much better performance under havy load as a single-threaded
> > >driver ... so their so called test results are very dubiuos and rigged.

> > >As Lincoln said:"One can cheat the poeple .. but only for a limited time"

> > >Armin

 
 
 

Linux vs. NT performance / Mindcraft results

Post by ?àDé? á??ò » Thu, 27 May 1999 04:00:00




>>1. In comp.arch.embedded it isn't so interesting...

>Nothing heard about 'embedded NT' ?  ... embedded LINUX ??

...embedded NT _SERVER_? ......embedded Linux _SERVER_? ...

at all - embedded _SERVER_? (... well... but it may be)

...embedded, not embedded... If we say 'server' (and "embedded" too),
don't wait echo "nt nt nt " or "linux linux linux", we'll hear another.

This both above for "clever and nice" intrface to user, no more...

Quote:>Armin

    At7
 
 
 

Linux vs. NT performance / Mindcraft results

Post by Rufus V. Smit » Fri, 28 May 1999 04:00:00



Quote:

>The test was paid for by Microsoft. A more interesting statistic would
>be the number of tests that came out favorably for the people who paid
>for the test results. Unfortunately, no one appears to want to pay for
>that study.

>                                              [sam]

ROTFLMAO!
 
 
 

Linux vs. NT performance / Mindcraft results

Post by r » Sat, 29 May 1999 04:00:00





>>1. In comp.arch.embedded it isn't so interesting...

>Nothing heard about 'embedded NT' ?

Hmm, dont know

Quote:>  ... embedded LINUX ??

Here is a good starting point (http://edge.fireplug.net).  A whole
router solution on a single floppy.  I think their links will get you
to other thin or embedded solutions.
Quote:

>>2.Yes. Linux has some problems with threads...

>No ... there are multi-threaded driver of RAID systems for LINUX.

>Why has Mindcraft choosen a single-threaded beta driver for the LINUX benchmark?

>Was it choosen by chance ??  I don't think so ...

>>3.Yes. Linux hasn't problems with drivers for newest devices...(no drivers -
>>no problems).

> ... there is some changing :-)

>Armin

 
 
 

1. * * * Mindcraft offer to re-run Linux vs NT test

According to an email from Microsoft today, Mindcraft have offered to
rerun the NT Server vs Linux benchmarks which were all over the press recently.

As yet the Linux Community (should read RedHat( have failed to respond.

Perhaps someone with contacts can get someone in RedHat to take note.

----------------------------------------------------------------------
OTHER NEWS
----------------------------------------------------------------------
UPDATE: MINDCRAFT OFFERS TO RETEST WINDOWS NT(r) SERVER VS. LINUX
http://www.microsoft.com/ntserver/nts/news/msnw/nt4vLinux.asp
Mindcraft recently published a study comparing the performance capabilities of L
inux to Windows NT Server. The Linux community has voiced several concerns about
 the test. Mindcraft has offered to rerun the tests and has met several of the r
equests of the Linux community. As of May 12, the Linux community has not offici
ally accepted Mindcraft's offer.
----------------------------------------------------------------------

Paul Gregg
--
Email pgregg at tibus.net |       CLUB24       | Email pgregg at nyx.net    |
Technical Director        |      INTERNET      | System Administrator       |
The Internet Business Ltd |    Free  Access    | Nyx Public Access Internet |
http://www.tibus.net      |  www.club24.co.uk  | http://www.nyx.net         |

2. timeout on a telnet session...

3. Smart Reseller questions Mindcraft's NT vs Linux tests

4. ppp trouble - part 2

5. Survey Results - Linux vs. NetWare vs. Windows NT

6. Driver Port Intel --> Alpha

7. Linux vs OS2 vs NT vs Win95 vs Multics vs PDP11 vs BSD geeks

8. A multicast problem

9. performance: linux/samba vs nt vs novell

10. Q: performance test results - Linux vs other 386x ?

11. performance of linux vs nt

12. JDK1.2.2 performance, Linux -vs- NT

13. Performance comparison of nt vs linux