Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×

Microsoft Challenges Linux community 639

AmirS writes "Microsoft are really pushing for the Mindcraft benchmark to be re-run, so much that they've put up a page about it. They say they've met all the requests of the community (seems like most have been met) and just require linux people to step forward for it. "
This discussion has been archived. No new comments can be posted.

Microsoft Challenges Linux community

Comments Filter:
  • by Anonymous Coward
    Um, I'm not a terribly techie guy, but I'd like to help make sure Linux wins this one legitimately.

    How can a non-programmer user help?

    (not anon, just no acct)
  • by Anonymous Coward
    Find a Linux project that interests you, and go ask them how you can help. Projects almost ALWAYS need help writing documentation, and who knows, you might learn a little programming while you're at it :)

    -- In numbers there is strength
  • by Anonymous Coward
    That's quite valid, I think. It's low-end in that it's a single processor box. If it's twice as fast, it doesn't matter all that much, since it gives both OSes a 2x speed increase. The problem is more with high-end features like SMP, me thinks.
  • by Anonymous Coward
    Hello All:

    From when I first started using Linux in 1996
    I've been told

    Linux is faster than NT
    Linux is faster than NT
    Linux is faster than NT

    a million times.

    So what's going on lately? Why all the news
    reporting the opposite??? First Mindcraft,
    now another one, PC mag:,675 5,2256617,00.html

    I still like Unix, but I don't know what to think
    anymore ;[


  • by Anonymous Coward
    Where is the url of this announcement please?
  • by Anonymous Coward
    Don't just measure raw throughput. Also measure latency. How long your customer waits to see a web site really does matter. If a web server can dish out terabytes of data per second, but does that by delaying some requests almost indefinitely, it sucks. I haven't benched throughput (and from Microsoft's claims, Microsoft may have pretty good throughput), but when I've compared the response time between IIS and Apache (not benched; just how long it takes pages to load on comperable servers), it's always seemed to me that Apache had lower latency.

    - pmitros on no sleep
  • by Anonymous Coward tml s/1999/mar/03296672.html
  • by Anonymous Coward
    Is this the best the Linux community can do? MS is willing to agree to your demands and this is the response they get...

    -"um, Linux seems faster but I don't have the numbers to back my statement". Nuff said.

    -"how about a 486 with 4 megabytes of RAM." Sure, thats fine for a personal home page, but how will a corporation serve a ecommerce site on that.

    -"my browser crashed at their site." Jeez, Netscape crashes on me at various sites as well. It's gotta be the sites fault. Actually, I would be very impressed if they could find specific HTML that would crash Netscape everytime. (Hmm, it doesn't carsh my netscape 4.08 or 3.01 or even 2.02.)

    -"they're probably going to optimize the kernel behind our back." I don't see how this is possible when the Linux reps will be there and other tests have backed up their findings using off te shelf software.

    -"MS is evil.", "MS sucks", "Linux rulez!". Uh,ok.

    Like a person suffering from a terminal disease, the Linux...actually, the \. community.. suffers from the following:
    Shock -- An anesthetized response to protect from pain. "No way!"
    Denial -- Not acknowledging the loss in effort to avoid pain. "No way! Mindcraft fudged the #'s"
    Anger -- Resentment at the loss or experience. "Mindcraft sucks, MS sucks"
    Depression -- A deep sadness or hopelessness.
    Acceptance -- Accepting the loss, learning and growing through the experience, and moving on. "Lets see where the weaknesses are and fix them"

    It seems only Linus has moved unto the last stage and realizes that the OS still needs to be improved and is willing to do that.
  • by Anonymous Coward
    So M$ is taking an INFERIOR operating system [Linux], that HARDLY has ANY MARKETING resources and is created by a BUNCH OF GEEKS serious?

    Now they want someone representing that operating system to come and have it compete against their SUPERIOR, SMARTLY MARKETED and by PROFESSIONALS created operating system [Windows NT]?

    Obviously they would have no reason to worry about any competition from this pathetic community attempt! What would it be then? I'm clueless!



  • I don't really think NT is a better OS. However I suspect that even if I was proved wrong by empirical data, the majority of the Slashdot community wouldn't believe it. They would deny it, and call it flawed. It almost reminds me of religious fundamentalists. I am a Linux user, and have been for years. I love Linux. I don't however make the assumption that it has no flaws, and is simply the best OS _ever_ no matter what anyone says.
  • by Anonymous Coward
    My feelings were pretty much the same as yours. Just a couple other things I noticed...

    "No centralized security - users must manually synchronize user accounts across servers" -- That's odd. My ISP, running Unix, has two user machines, and I get almost exactly the same environment no matter which one I log onto. I guess Microsoft never heard of NIS and NFS, huh?

    "More prone to security bugs" -- I just had to throw this one in because it's such a perfect example of MS's unsubstantiated attacks. As we say on USENET, "Cite? Cite?"

    Linux: "Hundreds of available applications" vs NT: "Over 8,000 Windows NT compatible applications available" -- Hey, MS, games don't count. ;-) But seriously, I suspect that a lot of these applications exist specifically to add functionality to WinNT - functionality which already exists in Unices of any flavor. (X servers, for example.)

    Linux: "Need highly trained system administrators - usually require developer-level skills" vs NT: "350K Microsoft Trained Professionals" and "160K Microsoft Certified Engineers" and "Integrated platform built around ease of use" and "Wizards to simplify complicated tasks" -- I don't get it. If NT is so easy to use and has all these great wizards, why do you need hundreds of thousands of MCPs? And aren't "Microsoft Certified Engineers" supposed to _have_ "developer-level skills"?
  • by Anonymous Coward
    From the ones that are allready up there, they are kicking linux's ass.

    But I know first hand, that Linux is faster.

    At my work, we have an in-office web server (as aposed to having somebody provide it for us). We used to run Windows NT 4 Enterprise server on a single CPU PII 450mhz, with 256meg ram. We had a licensed NT admin set it up, configure it, tweek it.

    Then after a long talk with a linux zealot co-worker, we managed to switch it over to linux, because the money wasn't there to upgrade the bandwidth or hardware (didn't really need it anyway). Anyway, just from surfing the web page from home, Linux helped it ALOT (it would process connection requests faster.... and it was just all around faster).

    Wish i knew more detail on how it happend, but this I guess is a Linux success story :).
  • by Anonymous Coward
    Lets sue the bastards! This is unacceptiable!

    They say that Linux supports only 2gig of ram.
    Then they say that NT supports 2gig user+2gig kernel out of the box.


    2gig user+2gig kernel is exactly what Linux supports! That is only 2gigs of ram.

    Get your fucking facts straight!

    Furthermore, Linux supports 64bit processors.. The RAM limit is a hardware limitataion (with Linux unwilling to adopt intel's stupid hack). I've booted ultrapenguin on a starfire with 8cpus and 16GB of ram.

    This kind of shit can not be tolerated. If MS published this kind of BS about Novell or SUN, they'd have their ass in court so fast MS's head would spin.

    MS has gone full tilt against Linux. Yesturday my local LIB system has begun removing Linux from 150 public access computers because in two weeks someone from the Gates foundataion is coming to offer them a grant for 50 more computers. A person from the GF told the lib director that she'd have no change of getting the grant if they kept the Linux. Since they already had windows licences for all their boxes, off comes the Linux.

    Bah! This all is very upsetting..

    I think I'll go kill myself now.
  • by Anonymous Coward
    certainly it has many more features, but it doesn't have more bugs. in fact, IIS4 is remarkably stable if you use it with MTS.

    Still not as stable as Linux/Apache/zeus/etc combo or Solaris/Apache/zeus/etc. wtf has a transaction server got to do with serving webpages anyhow?

    $1000 is one day's work for a development team. if over a project, using NT's superior tools saves 2 days, the choice of NT has paid for itself. (before anyone flames me over "superior tools", please tell me the linux equivalents of MTS, MSMQ and DCOM)

    Yeah but you get shite VB code monkeys for that, who need point and click because they can't code. I watched a Java consultant build a prototype EJB app in 8 hours yesterday and that included 2 hours explaining to the windows based developers how UI's work! He even managed to explain how the EJBs worked to me while he was doing it (These are guys from Silverstream, which is of course available for Solaris, and hopefully linux RSN)

    Also there is already Oracles Application Server out in beta that will blow and IIS application server away. Then there is zope, PHP, about 3 or high quality IDE's for developing Java, Perl, etc.

    There is actually already healthy competition. Given a fraction of the cost of an NT/IIS license and accompanying dev tools and apps I could have the same power and functionality and even a prettier interface. I would also perform the same job faster on less hardware.

    Then theres messaging - MQseries for AS/400 and other IBM connectivity, plenty of Transaction and message queuing software available already. Oracle and DB2 application servers provide most if not all of the features you need.

    Which server has more server side application options? Apache with it's module interface is superior to IIS and ISAPI. For example, rewrite and perl or python are much more powerful than ASP and jscript/Visual Basic

    ASP has complete integration with COM and can do anything a COM object can do. This approach is markedly superior to text processing languages.

    ASP is available for Linux as is equivilents such as php and zope - all the power but choice and stability too! besides COM is a very limited OO architecture and is no match for the raw power of Perl or C++.

    This same advantage is shared by CORBA and EJB tools, which again, linux doesn't have.

    er.. I think you'll find both on Linux and other Unices. GNOME is a corba based solution. Oracle App server among others of which there is a wide choice support both corba and ejb but also have perl/c/c++ support.

    Don't forget that to actually develop any thing worthwhile on NT/IIS (like database access and ecommerce) you have to puchase additional and sometimes very expensive tools.

    see above. the tools rapidly pay for themselves, then save significant cash. also, the cost of OS and tools is a very small part of the budget on major projects.

    Only in a microsoft shop. It takes vb programmers longer to knock out bodged junk than decent c/per/java programmers can build a decent app using widely available tools.

    The budget is far smaller when using open source, greatly reduced os costs, greatly reduced tools costs, similar application costs, cheaper hardware due to the ability to scale up thru SMP, clusters, and even architechures.

    Then you have the fact that you can move your entirely application and its tools up through the UNIX's available without having to recode more than a few lines.

    NT can't scale beyond 2 x 8 processor intel boxes - which cost far more than 16 x 2 processors but with out the redundancy. Or you could cluster multiple multiprocessor alphas or sparcs. Aaron (TheJackal)

  • by Anonymous Coward
    First, they're reporting it because it's news -- ie, it's different :)

    Second, this story has grown 'legs'. The huge response against the results was something journalists could write about!

    Third, there were extra bits to sniff out. Most people looked at the Mindcraft with Netscape or IE and missed the very tiny text saying "Test performed for Microsoft" at first. (Lynx-users, of course, saw it full-sized :)). The media enjoy a story where they can start digging, and the attempts to downplay MS involvement were just enough to interest the journos, who then learnt about Mindsoft's past benchmarking weirdness.

    So there's one reason why this story is still going on: it could be milked!

    One repeated comment you'll have seen is "it depends what you want to measure". Various alternative results have been posted both before and after (the difficulty is getting companies to provide numbers: most of us will know someone who said, "Well, internally, we found..." and then added, "But no, I can't say that, I'd need permission"). What you do hear, time and time again, is that for real "low end" machines, such as 386s and 486s, you can install Linux and it will _zoom_ along doing its job (webserving or email, often). You can't test _that_ versus NT cos you can't get NT onto the typical 386 or 486. So on those machines Linux is indeed incontrovertibly faster and won't crash.

    You'll notice that MS's idea of "low end" is beyond what you'll usually find in the company closet, In fact, Halloween 1 [] (scroll down a page or two from that anchor) explicitly mentions its use in this respect as a big problem for MS. On that note, if you're new to the more political issues surrounding Linux, check out the Halloween documents (so-called due to their leak date). They're internal MS assessments of Linux and free software. They're now well-read by Linux and free software people (you can tell how well-read by the fact that "OSS" for open-source software comes from those documents and everyone knows that abbreviation now! One thing that caught particular attention was the sections on "How MS should combat Linux", with such delightful admissions that FUD won't work. He didn't even have to explain what he meant by it, it was assumed that all MS types would know what it was... People have been expecting MS to try _something_ since then (and before). Other tactics for your viewing edification are legal threats (!) and "embrace and extend" open protocols, ie make MS the standard and everything else measure up to it. Actually, there's a flavour of that to this affair: suddenly "low end" becomes that ridiculous machine, and anyone who wants to use MS had better have that or buy a new one.

    This turned a bit long, but I hope it helped!

  • by Anonymous Coward on Thursday May 13, 1999 @05:54AM (#1893969)
    I'm guessing this one is mostly for the bigshots writing Apache, Samba and the kernel who know how to optimize their own code better than anyone else.

    However, there is a very large demand for non-programmers in the free software community. There is often a significant shortage of documentors, web masters, etc. on free software projects. If you want to write documentation for software, that would help more than you can imagine. If you want to make sure that web sites for programs have screenshots and other basic info, and are hooked up to freshmeat, that could also a big difference.
  • by Anonymous Coward on Thursday May 13, 1999 @07:07AM (#1893970)
    I liked this one most:
    Provides source code to allow developers to deviate from standard distribution.
    They got it! That's the sole purpose!
  • by Anonymous Coward on Thursday May 13, 1999 @05:30AM (#1893971)
    Like Zeus that is reportedly 3-20 times faster than Apache! Since we're pitting Linux vs NT, we should be able to use any webserver we want.
  • by Anonymous Coward on Thursday May 13, 1999 @05:38AM (#1893972)
    That page hurts. We are going to have to do something about this. There may be some FUD there, but there is a lot of truth too.

  • by Anonymous Coward on Thursday May 13, 1999 @05:51AM (#1893973)
    We've got Microsoft's list of selling points now. Address them one by one, and we've got a roadmap on what to improve.

    A few of them caught my eye as being near lies where they compare cost of ownership, etc against UNIX, not Linux. Tricky those Microsoftians.

    The most increidble selling point that they missed is the bulletproof nature of Linux compared to NT, and the work:crash ratio. NT is an immature product compared to the years of well-tested software that Linux has had to build upon.

    Wish we had numbers like the # of man-hours and such or bugs fixed to show.
  • by Anonymous Coward on Thursday May 13, 1999 @05:39AM (#1893974)
    Have a look at some of the claims on there

    "NO back compatiblity for a.out binarys"

    "No international support" when KDE supports more countries than windoze.

    "Poor support for java"

    "MORE PRONE TO SECURITY BREACHES".. remember teardrop?
  • by Anonymous Coward on Thursday May 13, 1999 @05:49AM (#1893975)
    ...that Linux really could lose.

    Calling stuff "FUD" won't make the challenge go away, nor will it make the _real_ technical problems of Linux go away.

    Linux still has a _long_ way to go with SMP scalability. Linux still doesn't have a journaling FS. Linux still doesn't have an LVM. Linux still doesn't have USB support. Linux still doesn't have a good, comprehensive administration GUI. [1] And unless you yourself are trying to help make these capabilities come to the point of production releases, you aren't helping at all.

    The best way to approach this challenge is to say "No, Linux doesn't quite measure up yet, but I'm personally doing everything I can to make it so it does." (Posting to /. does not count as "doing" :-)

    -- Citizen for the responsible promotion of Linux
    [1] All these capabilities exist in alpha form, but calling them production releases would be a LARGE stretch.
  • by Anonymous Coward on Thursday May 13, 1999 @06:40AM (#1893976)
    How about a third option.

    Why can't Slashdot, LinuxToday, or one of the distro companies prepare a set of test criteria that Linux performs better on than NT and challenge _Microsoft_? Let's think outside the box a little and quit letting Microsoft set all the rules of the game.
  • The original Anonymous Coward wrote:

    What exactly are Linux's strong points? It's not user friendly or easy to setup, has few apps, a chaotic development, is not all that fast (even *BSDs are faster). The only thing going for it it the fact that it's not Microsoft. Face it guys, Linux has been around for 8 years and hasn't progressed very far. It's a hackjob, a makeshift OS for the Microsoft haters. QED

    First off, blind assertion does not equal truth. If it is truth, than it flies in the face of the experiences of most people here, so some references or examples are pretty much required. If I were to assert "The sky is blue", few people would argue, if I were to assert "The sky is pink", I would need to support that statement, or it will be dismissed out of hand.

    Secondly, the original poster's assertions are either false, or so poorly defined that they couldn't be called "truth". It's not "user friendly"? What do you mean by "user friendly", I find Linux very user friendly, since it allows adminstrator-type users to access everything while restricting normal users from demolishing their system trying to install a pretty screensaver. I have set up many Windows 95/98/NT and Linux boxes, and I find RedHat much easier to install and configure than Windows.

    "Has few apps"? Have you looked at Freshmeat.Net [] lately? I don't know what you call it, but I don't call that "few". A "chaotic development"? Linux development is carefully managed and delegated. Microsoft is rumoured not to even let their programmers have full access to the code of the program they are working on.

    The blanket assertion that "even *BSDs are faster" is flawed on many levels. The tone is along the lines of "even this slow thing is faster than what you like", when most people consider the BSD's to be fast. I mean, "even horses run faster than you", what kind of comparison is that? It also isn't true without that tone. From everything I've heard, OpenBSD and NetBSD are generally slower than Linux. FreeBSD is faster only for certain situations, and only on the Intel platform. If I'm wrong, show me real references.

    "The only thing going for it is the fact that its not Microsoft". No, it also has "it's a fast, stable, general purpose operating system that works incredibly as a server and darn good as a desktop system". Also, "It's Free, both in the speech and in the beer sense!". These are big things going for it in many peoples books.

    "Linux has been around 8 years and hasn't progressed very far"? In 1991, Linux was barely more than an idea, in 1992, it still didn't know what SCSI or Ethernet were. Now it's a full blown operating system competing tooth and nail with megacorporations for being the platform of choice in the server room. I'd say it's progressed very very far.

    "It's a hackjob, a makeshift OS for the Microsoft haters", I don't even know what this means, much less how to respond to it.

    "QED", Latin for quod erat demonstrandum which was to be demonstrated. This being here means either the poster has no idea when to use "QED", or the entire goal of the post was not to answer the question "What exactly are Linux's strong points?" but to demonstrate that it is a "hackjob, a makeshift OS for Microsoft haters". Not only is that a pretty silly goal for a post, but it is a failure, because the post demonstrates nothing of the sort.

    In all, I think the post solidly deserved its -1 score. (No I didn't moderate it, otherwise I wouldn't be able to post this).
  • by Eric S. Smith ( 162 ) on Thursday May 13, 1999 @08:17AM (#1893981) Homepage

    As a registered [] member [] of the community, I thought I'd take a few shots at the MS "challenge."

    First, in the sloppy writing department:

    The Linux community has asked Mindcraft:
    • To configure and tune the servers themselves.
    • To be present to ensure that the tests weren't rigged.

    Well, uh, no: the consesnus is that when Mindcraft configures and tunes the servers (or not, as the case may be...), things go badly for Linux. Feelings on the second point seem to be the same.

    What MS mean to write, of course, is that Linux people want to configure, tune, and be present. I'm sure that the sloppy writing isn't intended to muddle the issue, since it's sort of clarified a paragraph or two down.

    Looking at their comparison chart, I note that they claim Windows has turned in the "best" scores on some benchmarks, while also noting that no Linux results exist. Winner by default, I guess?

    On Linux, it's "easy to gain root access...". But, they say, on NT:

    System services run in a secure context providing higher levels of security for multi-user services

    Does that mean that this exploit [] no longer works?

    Here's a nearly incomprehensible complaint about Linux:

    Low degree of integration increases costs and technical risk

    Melissa shows what costs and technical risks come with "integrating" stuff to the extent that MS wants to, but I'm not entirely sure what they mean by the word in this context.

    In the damning faint praise department, MS graciously admits that there are "hundreds" of applications available for Linux. Call me crazy, but Unix is, er, "several years" old -- I'm pretty sure there are more than hundreds of useful programs available (whether they're "applications" or not is not terribly relevant, if you ask me). Even if there are only hundreds, well, a comparison of quality, rather than quantity, would be more telling, I think.

    Another Linux failing, they say:

    No formalized field training

    I'm sure I don't know what that means. Organizations like The Learning Tree have Linux courses, and there have been a couple of certification programs announced (if inchoate).

    More Linux evil:

    Need highly trained system administrators - usually require developer-level skills

    Or, you could just give the job to some random person and let him/her peruse the manuals. Things wouldn't turn out any worse than they would if the person were told to run NT instead. The fact is that a Gooey WimpyWYG PointyClick screen doesn't change the fact that administering a computer well (let alone a network) requires skill, intelligence, dedication, and plenty of learning. No "Wizard" will get around this fact.

    Administrators are required to re-link and reload kernel to add features to OS.

    Uh, well, maybe. But you do have to "install service packs" on NT, which comes to the same thing in the end -- downtime while the admin does something that, if it doesn't work, will result in Bad Things happening until it gets straightened out.

    Most configuration settings require editing of text-based files

    Oddly enough, they forget the corresponding item on the NT side: "Most config. settings require editing of binary files." Or, rather, one (the Registry), and if you screw it over, God help you. At least the OS keeps a couple of backups by default.

    Here's one of their Big Awesome NT Features:

    Scriptable administration for automated local and remote management

    Unix is Home of the Script. That's all I have to say about that.

    NT feature:

    OS services and applications designed to integrated and work together

    Melissa. Not all rosy.

    Linux liability:

    End users forced to integrate...

    Nope, I'd say MS is the master of forcing people to integrate. (Yes, that was an out-of-context quotation followed by a cheap shot).

    NT feature:

    Over $2 Billion in R&D spending by Microsoft...

    And you know who's paying for that -- look at the prices of their OS and applications (particularly the proposed prices for the various Office 2000 flavours).

    And then they sum up. It's crapola in the best tradition of election campaigns, such as the one I'm currently enduring here in Ontario. Some highlights:

    Although the Linux community is focusing on the messenger and not the message...

    Well, when you notice that the messenger is full of shit, you don't tend to pay much heed to what's being said, now do you? The test was flawed (arguably fatally), so there's little point considering the results.

    Now it's time for the Linux community to demonstrate the real performance and scalability capabilities of Linux, or withdraw their criticisms of the initial Mindcraft report.

    No, Beavis, it's not. Even if no Linux person steps forward with brilliant test results in response to this "challenge," the fact remains that the original tests (and thus the original report) deserve the criticism they've received. This statement is about as valid as an assertion that since we have trouble treating cancer, we musn't go around saying how bad it is.

  • by davie ( 191 ) on Thursday May 13, 1999 @06:14AM (#1893982) Journal

    Nevermind the bogus claims this page is making about NT's superiority over Linux (Linux inherited UNIX' weak security because of buffer exploits? Like NT doesn't suffer from the same weakness?)--the fact remains that Mindcraft/Microsoft tried to pull a fast one and they were caught with their pants around their collective ankles. This challenge [] is nothing but kicking dust in the air until Microsoft and Mindcraft admit that they set the tests up to show Linux in a poor light. I am not willing to give Microsoft a pass on this one, and participating in their benchmark without forcing them to acknowledge the real reason for our dissatisfaction, instead of dismissing it as "attacking the messenger," would do a disservice to the Linux community. This would be like the archetypal battered wife returning to her abusive husband because he promises to clean up his act.

    It should be noted that one of the side effects of the Mindcraft benchmarks was that some very effective optimizations for Apache were identified and, last I read, were going to be submitted to the Apache group. I imagine the benchmark rules will prohibit their use in the proposed benchmark.

    The Linux community should respond in kind with a challenge to Microsoft admit that the first benchmark was a sham, that Mindcraft lied about the extent of their efforts to find help tuning Linux and Apache, and that Linux and Apache were de-tuned on the test machines. Last, but not least, they should fire Mindcraft.

  • by scottm ( 288 ) on Thursday May 13, 1999 @08:53AM (#1893984)
    Let's take a look at this honestly. I think, when pressed, few people would contend that Linux can scale as well as NT when price is not an option. (As pointed out by a previous poster, so what? Use Solaris if you need that enterprise level performance now!).

    Linux has an advantage in a few key areas:
    • Entry level-Mid duty servers
    • Adaptability [range of use]
    • Price [don't push it; no one looking at enterprise level solutions cares]
    • Lots of others that I'm not going to address here (:

    So let's look at these points and see how we might take advantage of them.

    Entry-level / Mid-duty Servers:
    Linux may well win this section of the benchmark. If not, it will be close. The addition of NT clients to the file serving test makes the test a bit more fair. Let's be a little reasonable about the test though and measure more than throughput. Request lag and reliability should be measured... Let's stick both boxes in a closet for a month serving some randomized requests and see who comes out alive (:... Should we use Apache? I don't know. If we just want to measure throughput or pure number of connections maybe not; so don't! Why do we have to?

    Adaptability / Range of use:
    Again, Linux rules here. Show me an NT box that can serve mail, web requests, smb traffic, ftp, etc and run on a PII/256mb... Now turn it into a firewall as well (: Push these points!!!!

    What to say? How about we just include the cost of the solution with the benchmark (:

    Now my next point. Let's do the test. Let's accept the results. And then lets come back in 6 months with a better product! If linux gets creamed somewhere, fix it and test again. Show how quickly linux can adapt and repair itself. Hopefully NT will have been slaughtered in some category as well; I'll bet it won't change in 6 months.

    Anyways. I'm getting sick of "FUD! FUD! FUD!" every time we see criticism. Let's take that criticism and use it!

    My $0.02.
  • This is from iCab, which generates error reports :)
    Altogether 35 errors found. Only 25 errors are listed below.
    Warning (1/1): is missing.
    Warning (22/1): The attribute "TOPMARGIN" is not defined for the tag .
    Warning (22/1): The attribute "LEFTMARGIN" is not defined for the tag .
    Warning (78/1): In the tag the value of the attribute "WIDTH" must be quoted.
    Warning (78/75): In the tag the value of the attribute "WIDTH" must be quoted.
    Error (87/48): In the tag the attribute "BORDER" is not allowed.
    Warning (87/48): The attribute "WIDTH" is not defined for the tag .
    Warning (87/48): The attribute "HEIGHT" is not defined for the tag .
    Warning (95/197): In the tag
    the value of the attribute "ONMOUSEDOWN" must be quoted.
    Warning (95/197): In the tag
    the value of the attribute "ONCLICK" must be quoted.
    Warning (95/586): In the tag
    the value of the attribute "ONMOUSEDOWN" must be quoted.
    Warning (95/586): In the tag
    the value of the attribute "ONCLICK" must be quoted.
    Warning (95/772): In the tag the value of the attribute "HREF" must be quoted.
    Warning (96/38): In the tag
    the value of the attribute "ONMOUSEDOWN" must be quoted.
    Warning (96/38): In the tag
    the value of the attribute "ONCLICK" must be quoted.
    Warning (96/206): In the tag the value of the attribute "HREF" must be quoted.
    Warning (97/38): In the tag
    the value of the attribute "ONMOUSEDOWN" must be quoted.
    Warning (97/38): In the tag
    the value of the attribute "ONCLICK" must be quoted.
    Warning (97/210): In the tag the value of the attribute "HREF" must be quoted.
    Warning (98/38): In the tag
    the value of the attribute "ONMOUSEDOWN" must be quoted.
    Warning (98/38): In the tag
    the value of the attribute "ONCLICK" must be quoted.
    Warning (98/204): In the tag the value of the attribute "HREF" must be quoted.
    Warning (99/38): In the tag
    the value of the attribute "ONMOUSEDOWN" must be quoted.
    Warning (99/38): In the tag
    the value of the attribute "ONCLICK" must be quoted.
    Warning (99/211): In the tag the value of the attribute "HREF" must be quoted.
    Warning (100/38): In the tag
    the value of the attribute "ONMOUSEDOWN" must be quoted.
    Error (104/5): In the tag the attribute "WIDTH" must only contain absolute pixel values.
    Error (105/5): In the tag the attribute "WIDTH" must only contain absolute pixel values.
    Error (106/5): In the tag the attribute "WIDTH" must only contain absolute pixel values.
    Error (107/5): In the tag the attribute "WIDTH" must only contain absolute pixel values.
    Error (113/1): The start tag for can't be found.
    Error (114/24): In the tag the attribute "WIDTH" must only contain absolute pixel values.
    Error (119/1): The tag is not part of HTML 3.2.
    Error (120/1): The tag is not part of HTML 3.2.
    Error (128/1): The tag is not part of HTML 3.2.
    Error (129/5): The tag is not part of HTML 3.2.
    Error (142/1): must not contain block level tags like
    • .

    • Error (148/1): The start tag for can't be found.
      Error (148/8): The start tag for can't be found.
      Error (150/1): The tag is not part of HTML 3.2.
      Error (152/5): The tag is not part of HTML 3.2.
      Error (159/1): must not contain block level tags like
      • .

      • Error (165/8): The start tag for can't be found.
        Error (167/1): The tag is not part of HTML 3.2.
        Error (169/5): The tag is not part of HTML 3.2.
        Error (178/8): The start tag for can't be found.
        Error (180/1): The tag is not part of HTML 3.2.
        Error (182/5): The tag is not part of HTML 3.2.
        Error (184/8): The start tag for can't be found.
        Error (190/1): In the tag the attribute "WIDTH" must only contain absolute pixel values.
  • This is actually not so hard. I thought about it for a while and several things occurred to me.

    1) The benchmark should be run 3-way:
    Solaris x86
    Linux 2.2.8
    NT 4.0sp5

    2) The high-end server must use a gigabit NIC

    3) The low-end server must not be a PIII
    (Microsoft will almost certainly hack in
    optimizations to take advantage of KNI)

    4) A test of database-backed web performance is
    mandatory (use MySQL... heheheh...)

    5) The tests should be run in a neutral setting
    (maybe once each at VA and MS labs)

    With these demands met, it should be possible to get a fair test, and even if Linux does get clobbered on the high-end box, Solaris should not. Linux will destroy NT on the lower-end box. So two sets of numbers are produced:

    "Solaris outperforms NT4 in the enterprise"
    "Linux destroys NT4 for entry-level servers"

    which will be spun by Microsoft as

    "NT4 outperforms Linux in the enterprise "
    "NT4 offers better price/performance than Solaris"

    but no one will care and the matter can be put to rest. Microsoft won't be conclusively demolished (and hence will be unlikely to try legal means to suppress the results) but neither will Linux, and (if these demands are made, loudly and publicly) MS will have to rise to the challenge.

  • (Whoops, tab->space fires the submit key...)

    "Dear Microsoft,

    Thank you for pointing out some performance flaws in Linux's SMP implementation through your Mindcraft benchmark tests. We are making changes now and should be able to remove these bottlenecks.

    We also wish to thank you for the list of bullet items on . While we disagree with the interpretation of some of these items, some are legitimate weaknesses of our OS and we are addressing them or will be soon. In gratitude, we have come up with a list of bullet items that you may wish to consider addressing in your operating systems.

    1) Physical vs. logical drive locations (drive letters)
    2) File organization (/home vs. put it anywhere)
    3) Support for multiple operating system file systems
    4) Lack of applications included in distribution
    5) No built-in way to run progs on one machine and display on another
    6) Limited platform support

    Respond to combativeness with friendliness. It'll drive 'em nuts!
  • People seem to be think that this page must be fair because it includes a sampling of three different benchmarks (two in addition to the disputed Mindcraft one). But what about this nice benchmark [] from Smart Reseller. This is the article that includes the wonderful quote: "According to ZDLabs' results , each of the commercial Linux releases ate NT's lunch." Microsoft included benchmarks by the other ZD magazines -- why did they "forget" to include this one?
  • A lot of sites use IIS 3, not 4 because 4 is still very buggy

    certainly it has many more features, but it doesn't have more bugs. in fact, IIS4 is remarkably stable if you use it with MTS.

    Is it $1,000.00 per server better?

    $1000 is one day's work for a development team. if over a project, using NT's superior tools saves 2 days, the choice of NT has paid for itself. (before anyone flames me over "superior tools", please tell me the linux equivalents of MTS, MSMQ and DCOM)

    Which server has more server side application options? Apache with it's module interface is superior to IIS and ISAPI. For example, rewrite and perl or python are much more powerful than ASP and jscript/Visual Basic

    ASP has complete integration with COM and can do anything a COM object can do. This approach is markedly superior to text processing languages. This same advantage is shared by CORBA an EJB tools, which again, linux doesn't have.

    Don't forget that to actually develop any thing worthwhile on NT/IIS (like database access and ecommerce) you have to puchase additional and sometimes very expensive tools.

    see above. the tools rapidly pay for themselves, then save significant cash. also, the cost of OS and tools is a very small part of the budget on major projects.

  • Do you realize that small sites exist?

    yes, of course, and linux is probably quite suitable for small projects. however, it's not
    suitable for everything, as some of it's more zealous advocates seem to proclaim.

    That's because you are accustomed to using the Microsoft tools

    it's not ease so much as capability. certain features, such as message queuing, ORBs, &c simply don't work on linux (for now; but i'm not holding my breath). And where's the journalled filesystem, HA clustering, transaction monitoring, system partitioning?

    Do you work for a company where budgets are very easy to get?

    not anymore so than any other company, i'm sure, but we do fairly large projects.

  • by Matts ( 1628 ) on Thursday May 13, 1999 @06:28AM (#1894026) Homepage
    Wow - that was harsh... I guess the Microsoft battle is now firmly ON - big time. That's probably good news - it means they're a little bit scared.

    So let's try and address these points.

    Spec Web

    The Spec Web figures are generally put out by hardware and software manufacturers in cahoots to produce high figures (often using slightly modified server binaries). Linux has no hardware vendors who are yet big enough to produce these figures. VA are getting there but I don't know if they have plans to produce SpecWeb figures.

    The same goes for SAP and TPC figures.


    The kernel developers are aware of some issues here, since Solaris and Irix don't exhibit this behaviour. Also the clients have so far tended to be '95/98 clients. Where I work all the desktops are NT Wks (thousands of them).


    These figures use Apache, instead of Zeus. Let's see some real figures with Zeus before making judgement here. Apache is meant for complex systems developers who need flexibility, not raw speed.

    Also, WebBench's dynamic benchmarks cover ISAPI on IIS and CGI on Apache. Gee, that seems like a fair test eh? It's not, and until WebBench provides an apxs module it will continue to be unfair to Apache. Why not compile that ISAPI module on Zeus and see how it fairs? I think we know the answer.


    Microsoft are most scared of Linux's reliability (hence it's at the top of their non-performance list). OK, so no OEM guarantees Linux uptimes. Big deal. NT's 99.9% uptime guarantees are based on clustering solutions - not single servers. And these guarantees are expensive. NTFS is not a true journaling filesystem either, although they may be talking about a commercial filesystem that I'm not aware of.


    I think they're probably pretty close on this - although they still make some glaring mistakes - like Synchronous I/O - only on the driver mindcraft used. And if pthreads aren't kernel level threading I'll eat my shoes. Yummy. Oooo and NT has an integrated file cache... Linux has one of those too. Wow.


    I don't even have to touch this section. Wow... stunning marketing going on in there.


    Comparing to UNIX, not Linux.

    Ease of Use/Admin

    I just loved the bit about "Scriptable administration tools for automated local and remote administration". OK, so wsh is now available. How much is it used? What about non-automated remote administration? For most administrators of NT boxes, wsh and SMS just don't cut it - they have to walk to the box usually.

    Actually, I'm bored with breaking this down. What can I achieve - leave me to just use Linux and be successful with it, and not have to suffer reboots.


    perl -e 'print scalar reverse q(\)-: ,hacker Perl another Just)'
  • You choose the right tool for the right job. So you want high SMP scalability, journaling file systems, LVM, USB and a world-class object oriented UI? Choose OS/2 Warp 5 Server. It's not open source, it's not free - but it works incredibly well. If you want it for free but with lower quality or with some features removed, choose Linux. Later kernel versions will hopefully take care of this, but for now...
  • Linux is like a diesel, it just keeps on truckin' and can really haul some weight and deliver the goods in a safe manner. 500,000 miles before a rebuild.

    Microsoft can only deliver its ego 1/4 miles, then crashes at the finish line, and requires an engine overhaul upgrade. Requires highly volatile fuel to run. Its for the thrill seekers. People who like to turn shades of white and red when the fun ends and the bill strikes. No practical use by me, thank you!
  • by dattaway ( 3088 ) on Thursday May 13, 1999 @06:22AM (#1894055) Homepage Journal
    Here [] is a nice description of what 99% really means.

    99.97% means 15 minutes a week is lost during an unplanned reboot, while Linux has a 99.9998% uptime, or 10 minutes scheduled.

    99% is pathetic when it comes to reliability! I want to see that number approach 100 the way a mathematician would be proud!

    After seeing NT hiccup last night on a production line, I feel offended! Plastic extruders powered by hundreds of horsepower each, capable of generating 500,000 pounds of thrust are dangerous to be around when temperatures drop (or rise!) and pressures exceed the massive iron head. When a few others and I saw the Visual Basic program decide to change temperature values to just below melting point, I knew we could have major property damage.

    Its the most amazing thing to watch large, high speed machines when the operating system freezes. Things keep on running, but never get updated. The once coordinated efforts of energy shaping a new product causes scrap to pile up quickly.

    Imaginge a half megawatt at the mercy of a single operating system and you have an idea how I feel.
  • And VA Research has SPEC numbers on their site last I checked.

    But there aren't any SPECWeb results on the SPEC Web site [] (that's "SPEC" "Web site", not "SPECWeb" site; i.e., it's the Web site for the Standard Performance Evaluation Corporation).

    A search for "SPECweb" on VA Research's Web site turned up nothing; where did you find their SPECweb numbers? (SPECCPU numbers, say, aren't SPECweb numbers; the only "SPEC numbers" that count as a response to Microsoft's claim are SPECweb numbers....)

    (There are NT+IIS numbers on the SPEC site.)

  • I haven't heard back regarding whether or not the LinuxWorld show wants to host a bake-off, but it sounds like we should go ahead with this even if they don't.

    I'm typing this in one of those $0.38/minute internet booths in Chicago airport on the way to the Dayton Hamvention (ham radio conference). May have spotty net access this weekend. I'll be back on Monday night, call me Tuesday at 510-526-1165 or email if you want to discuss this issue.


    Bruce Perens

  • I noticed.. I just thought I'd' keep my mouth shut...

    I'm sure at first glance of the "error" they would simply flip-flop them regardless of which ever is correct...

    Actually it is an error, their percentages are realy off. :)

    Marques Johansson
  • I could spend all day discussing the claims which are merely misleading or questionable, so I'll stick to those which are outright lies, besides the four already mentioned.
    • "Administrators are required to re-link and reload kernel to add features to OS".
      Hello? Modules? And the "reloading" statement is especially galling in light of Windoze's frequent need to reboot after installing a bloody application!
    • Historically, in order to perform optimally, applications need to be recompiled when the OS is upgraded
      Utter crap. Can you imagine recompiling everything each time the patch level is incremented? And are you telling me that Windows 3.1 applications perform "optimally" under 95 or NT, when they actually work at all?
    Ah, and under the NT column, I think the following bullet nicely illustrates MS' overall "commitment to quality"...

    "Why don't we address the int'l and accessibility point?"

    Steve 'Nephtes' Freeland | Okay, so maybe I'm a tiny itty

  • I agree with you 100% (MS equivalent of 99.98%)
    Why do companies like VA Research in particular remain quiet when they have the most to lose since they actually sell linux boxes that come with quad processors and gigs of ram! look at this page: ml
    if those machines are going to be running linux, and at those prices, i don't think the price of an OS will matter much to a company who is buying hte box.

    VA, if you guys are even reading this, don't just do a benchmark to refute MS, cause thats just petty. Step up to the plate (cause damn it, its hot right now) and tell people what your machines actualy can do! if it really can't do what you seem to be selling it to do, then you may as well pack up and go home, cause this is the real deal.

    Well, i am sick of ranting right now, i got some stuff to do. my head hurts...

  • by Kiwi ( 5214 ) on Thursday May 13, 1999 @11:50AM (#1894091) Homepage Journal
    I have put up a web page with my views of this challenge, and a rebuttal to many of the perceived weaknesses of Linux Microsoft notes, here:

    As an aside, I think the expression "that is FUD" is a cop out. If a fact stated is inaccurate, say so, preferably with a link to support the fact. For example, when Microsoft claims that Linux does not have a distributed security model, the reponse should not be "That is FUD". The reponse should be "Yes it does. For example, NIS is a distributed security model that works wiht Linux." Ideally, a link to some NIS page, such as the NIS HOWTO, should be provided.

    If the fact in question is true, hey, that's great too. Sometimes, opensource developers need more focus to do the best work they can, and what better focus than a challenge from Microsoft itself!

    - Sam

  • ...and they screwed up. Mindcraft is not a serious company, and do not deserve another chance.

    Let Microsoft spew their FUD. From what I gather, it has changed from primarily being aimed at the management, and is now instead focused on *us*. Microsoft is trying to discredit the Linux community. Just read the link in the topic.

    To the unitiated it looks like the tests were 100% fair (and not rigged), and that a bunch of Linux-kiddies now cannot tolerate the results. Pretty smart though...

    Anyway, this article sums it up quite nice: 5-penguin.html?05-11
  • We don't have to rig our benchmarks to make it work!

    Linux runs faster on my K5-PR166 w/64 megs of RAM than it does on the dual PII-233 w/128 megs of RAM at work.

    Whether MS likes it or not, we're here and we're not going anywhere. It can't buy Linux or slash prices and run us out of business. Even if Red Hat, Caldera, SuSe, etc. go under, Linux will still be here. We don't have to keep adding flashy eye candy and use slick marketing and lies to sell our product.

  • by fireproof ( 6438 ) on Thursday May 13, 1999 @06:27AM (#1894109) Homepage
    Yes, it's possible for Linux to lose.

    The hardware chosen certainly won't give Linux an advantage, whether it be the four processor box, or the low end system (although it looks like Linux would have a better chance on the low end box). Apache sure isn't gonna kick butt as a high performace web server (which isn't what it was designed for anyway). Of course, I wouldn't expect Microsoft to pick hardware that would give an advantage to anything but NT.

    Unfortunately, the "Linux Community" has made a whole lot of noise about the original Mindcraft benchmark and its problems. I fear that in doing so we have played right into Microsoft's hands.

    This benchmark business has already made so much noise that I've had people who have a hard time figuring out the Start Menu in 9x/NT asking about Linux, NT, etc. Microsoft is certainly doing all they can do to make sure the world knows that the "Linux Community" has disparaged the original benchmark. Now, they're raisin' a ruckus because we haven't (seemingly) done much to respond to their challenge.

    I think (and I'm probably not 100% right, I may even be wrong) that the "Linux Community" only has two options now:
    1. Accept Microsoft's challenge (however we're supposed to do that) and have as many experts on tuning, etc., on hand as possible to make sure Linux gets as much of a fair shake is possible given the hardware.
    2. Do nothing and let Microsoft trumpet to the press that the "Linux Community" didn't have the balls to accept their challenge because Linux really was inferior to NT (which we know good and well it isn't).

    Seems to me that our best bet is to accept the challenge and take our beating like a man, if that's what it comes down to. Of course, if Linux loses, MS will plaster the numbers on every bulletin board and window on this side of the Milky Way, but at least it will be a well tuned Linux running against a well tuned NT, which is much more than can be said about the previous benchmark. Even if Linux does lose, it's not the end of the world for us, even though MS will work to make sure it is. They can't kill us, and what doesn't kill us will only make us stronger. If this turns out to be a failure for Linux, then we can learn from what went wrong and work to improve performance in whatever areas it is necessary to improve performance in.

    And yes, I intend to help, as soon as I'm competent enought to do so. In the mean time, I'm learning.

    Now for a question: I was under the impression that NT pretty much choked on more than two processors. Am I mistaken? I know we've got an NT box with two PII (233s, I think) and 128 megs of RAM and it's slower than molasses).

  • Linux lacks a Journaling file system - file system may not recover after unplanned downtime
    "unplanned downtime" might be an issue for Windo$e, but ...

    Windo$e has an Integrated file cache for faster access to commonly used files
    What's new about that? Linux has got that for years.

    It's easy to gain root access on Linux via poorly written applications
    They don't say, that this is only true for suid-root applications.

    26% less expensive to set up and integrate than UNIX
    They don't say, that it's less expensive, because M$ admins can only double-click setup.exe, and if they're not happy with the result (after reboot), all they can do is reinstall from scratch (hopefully the right OS :)

    Historically, in order to perform optimally, applications need to be recompiled when the OS is upgraded
    What is historic about compiling? That's a FEATURE. In M$ OSs you have to buy an upgrade of your software.

    Windo$e has Support for 24K devices - 15K with Logo
    M$ doesn't write drivers, they just put their Logo on the hardware. And if your machine crashes, M$ will blame it on your broken hardware.

    Windows has Support through partners and OEMs
    They don't say, that the partners and OEMs can't do anything without sources.

    M$ complains that: Most configuration settings require editing of text-based files
    What's wrong about that? Aren't M$ server admins able to do that, or it notepad unable to open large text files?

    M$ says: Provides source code to allow developers to deviate from standard distribution
    They don't tell us that this is clearly a feature, not a bug.

    just my $.2

    Raphael Wegmann

  • And another thing...
    What is that with calling PCWeek a neutral location and using them to audit the results? MS Ziff Davis is not a neutral party. Let's do it at IBM or better yet, Intel. Using another part of the Micros~1 marketing arm, Ziff, is unacceptable. I also don't like that they want to compare to Test #2 and not Test #1. After all, the Linux community attacked MindCrap because of Test #1. I feel that Micros~1 knows if the test is rerun, they can't jam it down Linux's throat if NT wins when the results show a drastic difference from Test #1 because that is what this is all about...Test #1 publicized results.
  • The excitment around Linux is overshadowing NT and let's not even talk about developer interest. What also is amazing is that Micros~1 has a OS that they spent hundreds of millions of dollars to develop and they are attacking something that a bunch of guys put together in their spare time. How can there really be any comparison, test or no test? Linux wins just because it can be used to replace NT in many cases. Plain and simple, but will the public see it this way or the way Micros~1 is painting it? I wish there was a Linux Fund (like the Java Fund) only that the $$ were used for maketing Linux in general. To dispell the damage Micro~1 can do with its PR lies and FUD.
  • by Harlequin ( 11000 ) on Thursday May 13, 1999 @09:20AM (#1894152)
    The problem is, we shouldn't have to rig the benchmarks (like MS) to win. I have no idea whether Linux or NT would/will win in the new Mindcraft tests, but it seems like setting up rigged benchmarks is sinking to the level of MS. To do a vaild test, you have to identify what you want to test and then develop the test from there. You can't design a fair test to produce the results you want to exhibit (unless, I guess, you're MS :). Right now, it seems like MS wants to test the performance of mid-high end servers, and it seems like the test they've developed will actually test that. While it may be true that this hardware configuration is skewed to make NT look better, that doesn't mean that it's not a valid test.

    Say NT does beat Linux in a fair test (read not the first Mindcraft one). That shows that NT is probably better (for now) on high end servers. While I'm sure MS will produce some FUD saying something like, "NT Server is xxx% faster than Linux!!!", really what would be true is that NT Server is xxx% faster on high end servers. I think that, as a comunity, we should accept that and not try and hide from what may be the truth. What we should be doing is working to change that truth. Help develop Linux SMP, the file system, USB support, or whatever, that way, next year (or whenever) these tests are run again on mid-high end SMP servers, Linux wins, and wins fairly.
  • It is not in Microsoft's interest to make software which runs well on a 200. Remember, these are the people who practically invented the upgrade cycle, and part of that is making each software release require better hardware. It keeps people in the mindset that this is just how the computer world works, and no one will question whether it needs to be that way. They don't just want to sell you a bridge, they want to sell you the same bridge every year till you die.

  • Actually, I think the biggest difference is the hardware. NT can be very horrid on marginally support hardware (which includes many 'servers' and HCL systems with the NT sticker). On the other hand, it seems to stay up better on Compaq, HP and other name brand stuff.
  • by IntlHarvester ( 11985 ) on Thursday May 13, 1999 @05:57AM (#1894161) Journal
    Yes, it is a bizarre strategy to say things like "Linux has yet to post SPECWeb results", when "Linux" is really nothing more than a mailing list and some FTP mirrors.

    I guess you can file this with the No Roadmap FUD - non-sensical to anyone who understands what Linux actually is.

    It would make more sense to post "RedHat has no roadmap" or "RedHat has yet to post SPECWeb results", because RedHat is actually an operating system vendor, who at least in theory competes with the big boys, and therefore is going to have to (at some time) start doing the same kinds of marketing.
  • by IntlHarvester ( 11985 ) on Thursday May 13, 1999 @06:06AM (#1894162) Journal
    MS says:
    Overall, 37% less expensive to set up and operate than UNIX.
    26% less expensive to set up and integrate than UNIX
    27% less expensive to administer than UNIX

    Note that this may all be true, for regular commercial Unix, but the difference for Linux is certainly not that large.

    I'd like to see a real (objective) comparison between the operating costs of NT versus Linux in various roles. Note that I wouldn't be suprised to see NT come out on top (even with the licences), because the most expensive part of any server is still the system admin.

    (If you've got a good Linux admin working at your site for the same pay as a good NT admin, count yourself very lucky!)
  • by arivanov ( 12034 ) on Thursday May 13, 1999 @05:59AM (#1894163) Homepage
    Can somebody with authority finally cut the f... etherstripping crap.

    NT, Solaris (if tuned so) and other systems with more than one interface answer randomly from one of them. Linux answers where called. So as a result a server with 4 ethers on the same subnet will be lower under linux than any of the Mind... ZD benchmarsk.

    If somebody needs that speed use a gigabit ether and a switch. Than we come to where we should be:
    NT - 460MB/s
    Solaris - 850MB/s or better
    Linux 2.2+ 900MB/s or better.

    So can somebody with authority suggest THIS test so this crap is finally over.

  • The "rules of the game (and the model used in setting up the evalution)" were not based on common business practice, they were specifically chosen by Microsoft, at Microsoft's lab, because they show the greatest performace differential in their favor.

    You can't pretend this is a fair setup. The "unbiased" hardware used in the first Mindcraft test is just as suspect as the rest of their test has proved to be.

    A more fair test would involve either Microsoft and the linux team (whoever that turns out to be) agreeing on a neutral configuration, or alternatively a number of tests on different configurations.

    If MS is choosing this battleground, why shouldn't linux testers choose one of their own?

  • The ugly truth is that the community, through forums like /., sling just as much FUD about NT as MS does about Linux. If NT's reliability is so much worse than Linux's, show us the numbers? If Linux performs so much better than NT, show us the numbers? Singular personal experience statistically means nothing. Accept the MS challenge, learn from the results whatever their outcome, and stop slinging FUD. It is time to put up or shut up.
  • If they really want Linux people to step up to the plate, they really ought to make their pages readable by browsers their audience are likely to be using!

    Actually, they are making their pages readable by browsers that the intended audience is using. The intended audience is not the Linux market, as it may seem. They know that posted a web page telling about how much better NT is and how it costs different and is supported differently isn't going to cause Linux supporters to change their mind.

    I perceive from that page that they consider Linux a threat. Otherwise they would have ignored the whole thing and just been content with the results that they paid for. But now it has backfired on them by creating lot's of good publicity for Linux and bad publicity for NT. So now they create a web page targeted for those how are considering Linux full of tables and tables of useless figures promoting NT, with a little blurb at the top fingering Linux developers.

    Did reading the piles of figures change you mind? It didn't do much too mine. Especially where they talks about paper MCSE's and how much more it cost to run a Unix server. Unfortunately, it failed to talk about actual software costs. NT probably isn't cheaper then Linux after you add the costs of extra hardware and licensing fees to the amount you saved by using paper MCSE's who you can afford to pay minimum wage. It makes my glad that I got out of the Microsoft market, to a market where employer's can afford to pay *me* what I'm worth, because they don't have to pay Microsoft.

  • Perhaps we, as a community, could organize an independent testing group (made up of Linux users, but no kernel hackers, Redhat/Caldera/SuSE/etc. employees, FSF guys, etc.) to issue this very challenge. Perhaps several categories: best $1,000 server, best $2,500 server, best $5,000 server, etc. Entries could be received from any operating system whose advocates wanted to play: Macs, Microsoft boxen, commercial Unices, turnkey solutions like Cobalt, etc.

    I'll volunteer my time to help coordinate this, though I don't have the financial resources to test any category beyond the best sub-$25 system. :)
  • by Husain ( 16864 ) on Thursday May 13, 1999 @06:50AM (#1894190)
    The Linux vendors have screwed up!. Ppl. writing and using linux for "fun" (most of us) don't have an interest (or the HW) to optimize it on midrange servers (4ways with 1GB range of RAM). That is why it is possible that NT would beat Linux (Just as would solaris x86 would if you compared the two).

    If there is someone to blame it is all the Linux HW manufactures who are selling 4way Xenon's they are the ppl. who should have proposed counter numbers on the first day of Microcraft benchmarks (no but they just sat there).

    Linux OS ppl (esp. RedHat ) should have done something similar. They could have shown that Linux is not the best in midrange x86 servers but you shouldn't be using 4way x86 there anyway since you can run a Dual Alpha or SPARC with better performance (using other UNIXes) at a lower price. Frankly (except for being a little too confident) there should be no blame on the Linux community (who are just users and developers on small machines not big server).

    ... I am worring RH is not as smart as I though they were...

  • Reading some of the posts here and the Microsoft Article leads me to question what the goals of the community are.

    Linux will not be able to dominate the world today in its current state or perhaps in any state. Let us accept that, live with it and perhaps smell a few roses while we're at it.

    What are Linux's strengths. Let us use those to get at Microsoft. (This is from my perspective as a poweruser and a suit.)

    * Linux is a kick butt entry level and departmental server right out of the box. It provides VERY decent connectivity to Windows '95 and NT boxes out of the box with Samba. There are FAR more business using low end NT and Novell servers than there are companies using Quad Xeons. This is also where Microsoft makes its real money.. selling low end NT servers to departmental and workgroup users. Heck, my office in Hong Kong still has a 166Mhz Pentium as its main file server for 20 users running NT. We do not see any need for a P2, let alone an SMP box.

    * Linux is a VERY good platform for providing Internet services to the same businesses. Again, with a little bit of tweaking, the same Linux box can turn into a mail, web and news server. Try getting the same server described above to run exchange and IIS. Not possible. (Ironically our Exchange 5.5 server runs on a P2-450 with 256MB RAM).

    Performance is not everything. The type of box MS has constructed is used perhaps only in the largest of the large enterprises. Would any of the target audience seriously consider running NT on a $100K box? No, chances are they would opt for a SUN or an RS/6000/HP type solution, which ran their application/solution well.

    Let us look at Microsoft's Annual Reports and see where they make money. Is it by selling NT on a Quad Xeon? Or is it by selling Workgroup / Departmental Servers with '95 clients and Office bundles?

    My suggestion.. fight the FUD, but do not get consumed by it. The market is a LOT more than what MS has made it out to be. And look on the bright side of things... Linux is not featured on MS's pages. Departmental Managers, Small VARs looking at that page would go .. "hey.. do I need 400Mbps performance? Do I need a Quad Xeon?" If their answer is NO.. then perhaps with some positive press that Linux has received they should be asking themselves the question, "Why should I pay a couple of thousand to MS for NT when I can get Linux and support for a couple of hundred dollars?"

    Ok .. time to go to bed.. excuse the rambling :)
  • by DGolden ( 17848 ) on Thursday May 13, 1999 @06:12AM (#1894199) Homepage Journal
    This FUD is designed to focus the community into a producing small group to answer all their challenges.

    DON'T fall for it. DON'T answer their challenge too directly. DON'T provide a focus for their attacks. If MS has a single target for their attack, then they can set the rules for the fight.

    Keep diffuse. Attack from everywhere at once.

    Let them swipe at a swarm of bees with a sword, until they get too tired to fight.

    If big businesses are interested in Linux, they should come on our terms. Don't bow to pressure to soften the open source stance. We'll be here regardless of what they do, if they want to join up and help, well and good. If they want to fight, eventually they'll lose.

  • NTFS provides a 64-bit file system which is capable of file sizes up to 264 (must larger than 2GB) ^^^^

    I was hoping I'd be the first to mention that one, I was beside myself laughing when I read it.

    You've got this big gun-ho web-page exagarating the size of Bills testicles and their complete superiority. And Microsoft Office 2000 can't even pick up a simple grammar error. (yeah, simple things - simple.. etc).


    I honestly think they are trying to set Linux up for a fall. If(/when depending on your viewpoint) Linux beats them in a benchmark. They can still claim a victory? Why?

    E a s e O f U s e.. (I hate the phrase too)

    If they win, they win. If they lose, all they have to do is say look; you have to hire the top-wiz-bang linux hard-core widget-builders to get Linux to outperform WinNT. And, they win.

  • by Mr. Piccolo ( 18045 ) on Thursday May 13, 1999 @09:34AM (#1894202) Homepage
    Now how about explaining why they're all false?

    1. NO back compatibility for a.out binaries

    Oh really. I suppose that's why the kernel has a "Support for a.out binaries" option in it.

    Plus, Slackware at least has the entire a.out compatibility libraries with it. I'm sure there are others.

    Frankly, I don't see why this point is relevant at all, since Solaris is ELF, the BSDs i believe are using or moving to ELF, etc...

    2. No international support

    And that's why X has Japanese and Cyrillic(sp? Russian) fonts. Plus you can use all those nice ISO-4digits-moredigits fonts to get all those funny little accented letters, should you need them.

    Oh, wait, they meant the OS has to understand Unicode. Just more "Our way or the highway" thinking.

    3. Poor support for Java []. Nuff said, except Sun themselves have endorsed them.


    Actually, that would be "less" since security breaches are fixed instantly (more or less) by the sheer number of coders out there. This is in contrast to Microsoft's model where it takes forever (in Internet terms) to get a fix.

    Yep, you know what come next after "They fight you..."
  • Here's an idea that may help us to fight back against the evil spectre of MS-HTML. Write webcrawler software that searches the web for MS-HTML masquerading as HTML. Build a database of noncompliant web sites, along with a list of e-mail addresses of the authors. E-mail the authors of the web sites to alert them to the problem, and provide solutions.

    Most people are unaware that the nonstandard HTML known as MS-HTML even exists, so if people are made aware of the problem they can fix it.

    "Proprietary standards" is a contradiction in terms.
  • You should all read Petereley's comments in LinuxWorld []. He believes that it is a bad idea to play their game on their turf, i.e. continue with the Mindcraft benchmarks.

    I have to say that I generally agree with this. I think it would be far better for the Linux community to do its own benchmarks using more typical hardware configurations and a realistic mix of applications.

    In other words, forget about competing with NT in an artificial playing arena. Benchmark Linux in a realworld scenario.

    This is something that Linux vendors like VA Research should be doing in conjunction with Caldera, RedHat, et al.
    Michael Dillon - E-mail:

  • Let's think outside the box a little and quit letting Microsoft set all the rules of the game.

    Precisely. That's how we've got where we are. There is no advantage in playing Microsoft's game. Even if Linux/Apache/Samba wins the benchmark (certainly not guaranteed, given known issues), MS will spin it that the only way you'd get that performance yourself (if you went Linux) would be if you hired Torvalds, Cox, etc to tune your systems for you. If we lose, even if the margin is tiny (compared to the original Mindshaft tests), MS will trumpet that as 'proof' of NT's superiority and continue to quote the first set of numbers. (We haven't heard the results of the second benchmark, have we?)

    If we simply don't show up, Microsoft can say what they like but folks out there will remember how they skewed the first benchmark, and knowing how trustworthy Microsoft is (ahem!), will as like as not say "hey, I don't blame them, why get screwed over a third time".

    Microsoft is running scared on this, they don't know how fight something they can't buy out or bury. Let's just keep them off balance, and ignore this particular challenge.

    Given Linux's ability to run on many different platforms, it might be interesting to spec out what configuration would deliver benchmark numbers an order of magnitude higher than anything NT is claiming. I doubt NT would even run on that hardware, and the hardware might cost more than the quad Xeon of this test. But so what? If Microsoft wants to get into a price/performance match with free software... Well, I don't think they'd really want to go there.

  • by Ellis-D ( 19919 ) on Thursday May 13, 1999 @06:26AM (#1894225) Homepage
    Well I have decided to do the /. supported Linux distro..If you want to help email me at or icq me at 1763538
    Hopefully we can get Rob to support it!
    "Windows 98 Second Edition works and players better than ever." -Microsoft's Home page on Win98SE.
  • by flimflam ( 21332 ) on Thursday May 13, 1999 @05:32AM (#1894232) Homepage
    Microsoft talks about the Linux community as if it's something monolithic like a competing company. I understand that that's probably the only way they know how to respond to something, and even if they did understand us it's still in their interest to denigrate linux I guess.

    It's mostly interesting to see how threatened they apparently are by linux -- the OS that just won't die!
  • by Eric Kidd ( 21408 ) on Thursday May 13, 1999 @05:51AM (#1894233) Homepage

    So Microsoft's prepared a big page with some truths, some nonsense and a few inaccuracies.

    The major Linux vendors needs to run some public price/performance benchmarks of their own through various respectable organizations. I bet Pacific High Tech can do something useful with clusering, for example.

    But overall, we should thank Microsoft for providing such valuable feedback. ;-)

  • Would it be cruel and unusual to throw in a few other OSes?

    I for one would like to check out how BSD networking compares

    Lets do one set of bench marks, using a variety of oses each tuned by some celeb/pit crew? let's see' what OSes run on INTEL? Perhaps it should even be made a annual event. (kind of like the olympics)
  • by digitaldaniel ( 24033 ) on Thursday May 13, 1999 @05:31AM (#1894246) Homepage
    Mindcraft has added a "low-end" configuration (single processor, 256MB).

    Wow, when you shift through all the FUD and M$ propaganda, this seems to be be the real jewel of the artical.

    While I believe Linux will probably be out performed on the absurd high end server, we have always voiced its superiority on lower end equipment (well not low end, but not this M$ box).

    But then again, this is M$ (regardless of ZD labs participation), and they could still pull something out of their A$%
  • by Hedonistic BOFH ( 27550 ) on Thursday May 13, 1999 @06:37AM (#1894260) Homepage

    Here's an interesting graph []. It may not be all that scientific, but it gives a good idea.
  • Just an interesting aside, looks like they forgot to clean up some draft work.

    Pretty far down in the section:

    Integration of system services and applications to reduce complexity and management costs

    There is a 'ding' against linux:

    Open questions about internationalization, access by people with disabilities

    And then it looks like they meant to put their NT4 answer, but forgot to replace their draft 'placeholder':

    Why don't we address the int'l and accessibility point?

    Hrmm... why *don't* you?

    Just thought that was funny.

  • MS says the Linux community is slow to respond?
    Everyone remember the recent Salon article at
    http://w raft/index1.html []?

    Read it again to refresh your memory on how cooperative Mindcraft has been.

    Couple of quotes:

    "I've traded a couple of e-mails with Mindcraft people about this," says Alan Cox. "They seem solely intent on trying to re-create their existing pro-Microsoft results and hoping, by attaching some kind of 'Linux top mind' credibility to it, they can do more damage."

    "The whole thing has been fairly painful," says Torvalds. "Mostly because these people don't actually let us know what the hell they are doing. We've been offering to be on site to see what the hell is going on, but so far they've refused."

  • I work in the real world, too, on web-based applications. According to everything I've read, Linux stops scaling at about a dual processor machine with 512MB of RAM. Let's get some perspective: Dual PIII with 512MB is a huge amount of computing power. At the point that my application needs that much power, I'd want to look into (affordable) redundancy, which would mean adding another server.
  • Who would even dare to guarantee such a thing as a 99% uptime for NT? I wonder why the article never mentioned WHO guarantees it, that would be interesting...
  • This is the bottom line. No matter how many rigged benchmarks Microsoft or anyone comes up with, Linux is still working for the people that try it. I work for an almost all Linux ISP. Our only NT box is a web server. It takes twice the muscle to do half the job. Plus it is almost impossible to administer remotely (without laplink that is), has to reboot everytime you add an IP address, and everytime you do something new you need to drop big bucks on software.

    Bad benchmarks will slow down acceptance of Linux, but not stop it.

  • You make an excellent point. The wording seems to characterize "The Linux Community" as a competing corporation

    "The Linux Community, Inc." -- trademark it and reserve the domain now!
  • Wow! You can turn NT into BSD? It's got Berkeley Unix embedded within?

    I'll have to radically change my opinion on NT now...;^>

  • I don't really think NT is a better OS. However I suspect that even if I was proved wrong by empirical data, the majority of the Slashdot community wouldn't believe it. They would deny it, and call it flawed.

    I would deny such empirical evidence because I have seen the evidence myself. I run both NT and Linux, and work with access servers of both types. The Linux box does more with less hardware. It stays up longer.

    If NT were faster, I wouldn't need Microsoft telling me so. I'd have seen it myself. For at least some (I suspect most) of us Slashdotters, we don't like Linux because it's cool, hip, or countercultural. We like Linux because it works better than anything else for most of what we need to do.

    Show me numbers telling me that NT outperforms Linux in reliability or speed, under all but pathological cases (and the Mindcraft test is pathological--perverted beyond any real-world applicability), and I will deny them. Unlike a fundamentalist, I do not deny them by faith. Like a true hacker, I deny them by experience.

    All the uptime figures in the world don't change the fact that I have to reboot my NT machine every week while my Linux box stays up until the chips blow. All the feature set listings don't change the fact that I can easily administer my Linux machine over a 28.8 modem, saving me an hour in driving time. All the performance numbers don't change the fact that my sizeable network now relies on a base Pentium with Red Hat as the augmented FTP server, where a P2/300 would be minimal for a similar NT solution.

    When statistics deny reality, I will deny statistics over reality any day. When NT stays up and does its job quickly, I will be duly impressed. I won't need a page of figures to tell me, though; just an upgrade.

  • Amusingly, Microsoft's mention of The Linux Community points to a Slashdot story. Smile, you're on NT Camera!

    You couldn't really call it "The Linux Community, Inc." without actually incorporating it. Doing that would piss of the hard-core Linux hackers (distros can be corporate, but Linux itself cannot). How about "The Linux Community, Uninc." (unincorporated)?

  • InfoWorld, Volume 21, Issue 19 (May 10, 1999) has a notice on page 14. The last line:

    "Look for InfoWorld's file and print benchmarks of NT 4.0 vs. Linux in next week's issue."

    The Linux community can't put together a benchmark that looks independant (whether it is or not) for the same reasons Microsoft can't: both are interested parties. Sometimes the best thing to do is wait for someone else (e.g. the trade rags) to do the job.

  • I'm no defender of Gates at all. But accusations that he isn't a programmer are unfair.

    Remember the Altair8080? Recognised as the first PC way back in the heady days of '72, Gates and Allen wrote a version of BASIC to allow users to program for it in a way other than flicking switches. Not to put too fine a point on it, this was quite a programming achievement at the time and this alone is enough in my mind, for the MS founders to be given at least a little respect.

    Was, not is. Bill Gates is not a programmer, just as O.J. Simpson is not a football player. They both were decades ago.

    Actually, it's kind of sad. Gates could have been an uberhacker, and was making large strides down that path. But it has been said that the greatest of the angels was the first to fall. The story of Microsoft would have made a great Shakespearian five-act tragedy.

    In my book, Gates earned a lot of respect as a programmer. He has since squandered every bit of it and gotten into some serious zorch debt with me. I respect him as a fellow human being; nothing more and nothing less.

  • by Izaak ( 31329 ) on Thursday May 13, 1999 @07:05AM (#1894285) Homepage Journal
    Benchmarks like this are a load of crap. It is like racing a dragster against a stationwagon. Sure, the dragster can go really fast in straight line... but it sucks gas, corners like a cow, and is expensive to maintain. Try to do anything useful with it (like go to the grocery store) and you will be going nowhere fast.

    So NT beats linux on a particual hardware config (4 CPUS, 4 ethernet cards) in a particular test (static web and SMB file serving). I am sure MS went to great lengths cramming support for this down in the NT kernel so they could get those results... but at what expense? Just like a dragster, they now have a machine that goes fast in a particular setting... but sucks even more for general use.

    Listen. We could no doubt hack the Linux kernel to do the same tricks as NT, do them better, and trounce all comers in these benchmarks... but why!? People don't buy servers to win benchnarks, they buy them to do real work. We need to cut through the FUD and remind everyone of that. We should continue to work on improving linux performance (SMP does need work), but not at the expense of flexibility or stability. We need to counter these slanted benchmarks with our own tests that more closely appoximate the real world demands placed on an enterprise server. This is where Linux mops the floor with NT.


  • If you work in the financial industry, time is worth a hell of a lot of money. Developers are expensive, and time to market can be worth millions (or whatever. A new tool for a trader could generate a significant edge worth millions). £1000 is nothing (especially as a large organisation can arrange discounts).

    As for the advantages of using component based software. Well this is hardly revolutionary. Our website would be a nightmare if we were using Perl. An unsupportable mess at best. And what would we do when we needed a transaction server? Distribution and replication of data? Perl has its place, but I'm not convinced that it scales for data intensive tasks. Which is not to say that NT does them well (it does them very badly in my experience).

    Java boy
  • Sorry, as someone who develops on both systems I have to respond to some of your points...

    A lot of sites use IIS 3, not 4 because 4 is still very buggy.

    A lot of sites may still use IIS 3, but it's because they have lazy admins, not because there's anything inherently wrong with IIS 4.

    Just for the sake of argument, lets say NT/IIS4 is better at the upper end. Is it $1,000.00 per server better?

    Gotta agree with you there. What really gets me is that NT Server and NT Workstation are so similar (aside from some different registry tunings) you're basically paying for IIS - yet IIS is, according to Microsoft, "included free with NT Server"! Sheesh.

    Which server has more server side application options? Apache with it's module interface is superior to IIS and ISAPI. For example, rewrite and perl or python are much more powerful than ASP and jscript/Visual Basic.

    Sorry, I disagree. You can do anything in ASP that you can do in Perl and Python. ASP also grants you access to COM and ActiveX components so you can programmatically access server-side components developed in C++, etc.

    Don't forget that to actually develop any thing worthwhile on NT/IIS (like database access and ecommerce) you have to puchase additional and sometimes very expensive tools. Tools that are free or low cost for Linux.

    I can't comment on the e-commerce side (although I'm not convinced that any enterprise-ready e-commerce tools are available for Linux for free, either), but re: databases: for COUNT(*) less than 1e6, what's wrong with Access?

    As always, your mileage may vary... I think a lot of IIS's purported instability is due to incorrect setup of the underlying OS (why do people think that just because NT's graphical any monkey can set it up properly? :)

  • "Even though their requests have been met, the Linux community has not officially accepted Mindcraft's offer."

    I'm does a community of hundreds of thousands of people "officially" accept a challange? It seems MS knows how to attack a copeting organization, but they have no clue how to deal with a community.

    Certianly bashing the Linux community isn't going to help, becuase these are *users* being bashed. How can you hope to win over users by attacking them?

    A sales visit from a Microsoft rep must be interesting. "I see you have a Linux box. You must be a moron. I'll condescend to sell you this NT box instead, even though you will probably have to put an X here on the contract, as you clearly must be an illiterate. Please don't drool on my Armani shoes while you're signing..."
  • Most people missed the significance of IBM's announcement last month when they introduced a naked laptop. This was a direct assault on Microsoft's 100% monopoly market: the laptop. Do not expect Microsoft to go quietly into the night. They will fight back with everything they have to protect themselves.

    What will they do next? Reread the Halloween Paper. It details some of Microsoft's options. Don't be surprised if they file a lawsuit against some Open Source developers.
  • Did anyone else read the article in the current Business Week about Reinventing Microsoft. There's a passage in there about how Steve Ballmer had to dispatch 20 (yes twenty) field engineers who took several days to get acceptable performance out of a large company's NT installation.

    And they say it's easy to use.

  • 1) According to the page (put up Yesterday (12th)) The Linux community have been slow to respond. Jeez guys, it was only made known yesterday. 2) MS/Mindcraft definition of a "low-end system" 1 processor / 256 Mb of memory ????? And what will that processor be, pray tell ? Lets see NT run on a DX/2-66 with 16Mb. Now *that's* a low-end machine. A.
  • I've got quite a few thoughts on the Microsoft challenge. I don't have the technical background (yet) to determine whether particular configuration tweaks are fair or not, but I didn't get a degree in philosophy for nothing. I'm going to try to map out the challenge for strengths and weaknesses of claims and arguments Microsoft makes.

    In the first paragraph labeled "The Mindcraft Report", the author writes that PC Week and PC Magazine tests have corroborated Mindcraft's findings. If true, this is significant because it has been claimed that Mindcraft's study (and its study alone) contradicted prior studies. To show it's true, however, these questions (at least) must be answered:

    1) PC Week and PC Magazine tests must be replicating equivalent conditions (per each other and Mindcraft) to be said to "corroborate" the original study.
    2) Each set of tests must be "fair"; that is, they must not suffer from the same kinds of fatal flaws Mindcraft is accused of incorporating in their methodology.

    Furthermore, even if the two studies check out with the two above conditions, that alone does not validate the original study. The original methodological concerns first raised must be satisfied (as they presumably will be should this now-hyped "open benchmark" test take place.) In short, the two new sets of tests do NOT "prove" the first Mindcraft study; at best, they provide a bootstrap to give further credibility to a presumably pro-Microsoft (in terms of winning) verdict in the possible future Mindcraft open benchmark.

    Moving on, the Microsoft paper commits a subtle slur against the Linux community in intimating that the Linux experts are "dragging their feet" in responding to Mindcraft's new challenge (implied to be a quick, fair response.) Three things here: (1) Mindcraft submitted its original report on April 13. The open benchmark challenge was first issued May 4 and revised May 7. Given that the second "release" is the one that might be taken seriously, it's now been one week since that challenge went out after it took 3 weeks after the initial report to issue it. The "slow-to-respond" charge doesn't seem to take this proportionality into account. (2) Mindcraft is a single company, and this is work-related; they can do this "on-the-clock", as it were. Linux community experts are dispersed worldwide and by-and-large have jobs that demand their time and effort apart from their Linux roles. It's a major effort to collect all these people in one place for a conference planned months ahead, let alone a benchmark test in four weeks or so. (3) This charge is deliberately made because there is no way to decisively refute it. All it takes is one person answering late or refusing to participate, and one can paint the "community" as being recalcitrant. This is a barbed challenge, make no mistake.

    Now the "track record": take it one point at a time. The TPC-C part is, frankly, very weak; I think it was put there in the hopes nobody would check it. The actual study primarily measures two criteria: throughput and throughput/cost of system (as determined by an entire, integrated system.) In the link Microsoft provides, the TOP throughput number is 24328. However, the _10th_ best in pure throughput has a score of 48793, just over twice as much as the "top" MS solution! (Number 1, by the way, is a Sun Starfire system with a throughput number of 115,395!!) By and large, the throughput list is comprised of high-end UNIX flavors whose price tag keeps them off the lists Microsoft so proudly displays.

    In short, Microsoft's accomplishment here boils down to being the best cut-rate solution running on PC hardware there is. And the competition here is....? This is the whole point of the DOJ trial; MS just narrowed the field down until it hit its monopoly chip and then paraded the results. It is true that Linux vendors have yet to submit these kinds of benchmarks. Of course, what was the state of Linux vendors one year ago (how many, how successful, etc.)? I submit that the up-and-coming players today have been too busy trying to take the market by storm to worry overmuch about benchmarks.

    As per the SPECweb: MS makes the claim that they have the "best dual and quad processor results." Well, according to the single page they link to, they must mean that IIS5.0 on a HP Netserver 8000 beats Apache & IBM on HP Netfinity 5500/7000 for 2 and 4 processors. IIS4.0 doesn't do all that well, and HP 9000 with Zeus absolutely beats up on MS's results. The "best" claim is optimistic and near-sighted at best. Again, Linux isn't in this--yet--and I'd suggest the people to talk to are the HPs, the IBMs, and the Suns represented in this particular benchmark. (Again, Linux is NOT one company!! Actually, in light of this fact, the claim "Linux has yet to post SPECWeb results" is a little bizarre.)

    Re SAP: again, SAP has but recently made a Linux decision. (This entire process, by the by, is somewhat akin to the local bigwig claiming the new kid in town doesn't deserve respect because the old families haven't met him yet. The answer is both cases is, "Give it time.") As per the technical SAP evaluation (if you can find it), I'm punting on that one. If you're an expert, think of critiquing this claim as a module to plug into my larger argument.

    THE REST: I'm about to beg off because I've got other things pressing. However, I've got some remarks that I think will cover most of the remaining claims that MS makes. First of all, the NetBench and WebBench tests use results purely from the Mindcraft, PC Week, and PC Magazine tests. My above comments should be kept in mind when evaluating these numbers; furthermore, these other two studies are brand new. I think it is not at all a coincidence that the studies and this gauntlet are so close together. It doesn't quite smack of collusion, but it does suggest that the marketing folks over at Microsoft instigated this document to capitalize on the prima facie positive results. By the time solid critiques of these two studies (coming out on the same day, no less) can be made, the marketing machine will probably have moved on. This doesn't imply that the two new studies are flawed; I simply suggest that whether they're flawed or not won't ultimately matter in terms of that new god of mass media, "perception."

    The "Performance" section has three main flaws: (1) The points often don't match up against one another, (2) The lack of a centralized, bureaucratic command policy is always presented as a negative with a corresponding (false) positive always placed on integration and command decisions, and (3) many of the claims are either false or only true in the most trivial sense (see the security section, for example.)

    To conclude, a shrinkwrap blurb does not an argument make. Of the "track record", two sections are really meaningless, a third is likely so, and two others need to be evaluated before proper judgment can be passed. Of the "performance" criteria, almost everything is a comparision of one paradigm element to another (and as such aren't suitable for comparision).

    To response to Microsoft's "challenge", I think that the PC Week and PC Magazine tests need to be scrutinized. I think the Mindcraft test in June _might_ be a good response (although almighty tough to win in terms of PR), but it is VITAL to establish the significance of the test BEFORE it happens. I think it is a mistake to have a "wait-and-see" attitude; that is, it would be foolhardy to wait until you've "won" or "lost" before you say the test is meaningful or not. Finally, there is one part of the challenge that I think MUST be taken up: the response to MS' criteria for performance. That's an attempt to define the battlefield and absolutely MUST be countered.

    Haste makes waste when it comes to the first half; knee-jerk reactions to benchmarks studies or rushes to establish benchmarks (like the first three tests they cite) for the sake of having your hat in the ring are ultimately detrimental to my mind. But when it comes to the second half, "he who hesitates is lost." Think of it in terms of warfare. Microsoft is a cadre of armor-clad knights, the heavy infantry of medieval times. Linux advocates are hill warriors (the Scots, say). Hill fighters don't work well on plains, and strong horses are useless going up mountains. Where are we going to fight the battle?

    To do well, we must fight in the hills. Integration? Centralized control? No, no, no. Choice. Flexibility. Education. Openness. This is how we establish where the battle is fought and how ultimately the challenge is met and surpassed.
  • >>$1000 is one day's work for a development team. if over a project, using NT's superior tools saves 2 days, the choice of NT has paid for itself. (before anyone flames me over "superior tools", please tell me the linux equivalents of MTS, MSMQ and DCOM) >ASP has complete integration with COM and can do anything a COM object can do. This approach is markedly superior to text processing languages. This same advantage is shared by CORBA an EJB tools, which again, linux doesn't have.
    Corba is available for linux/python, it's called fnorb and has been around for at least a year. Python is not just a text processing language. Unlike VB and vbscript/jscript it is fully object oriented and can interface with C/C++ libraries if needed. Not just made to fit libs like MS does, but any C/C++ libs.

    Corba is multiplatform and at this stage far more robust than COM (I use Delphi and COM). Corba doesn't lock me into Windows specific tools.

    BTW, Pearl and Python have "standard" features that VB programmers can't even dream about.

    Now about MTS,MSMQ and DCOM.
    MTS... very good, but not unique, the Borland/Inprise Midas architecture can use it or CORBA, and can use corba on non MS platforms.

    MTS is just a another buzz word from MS thats been done on other platforms. You don't need MTS to create you own transaction management system, though MTS is supposed to make it easier.

    DCOM.. what corba used to be. There is also RMI if you use Java, with a much more robust communication process.

    MSMQ, don't know what that is, sorry.

    COM,DCOM,ActiveX.. all add processor overhead because of the requirement to use virtual methods and unknown types. Makes it easier to implement, but slower in setup and intial execution.

    Advice, next time you call a language just a 'Text processing" language, make sure you know what you are talking about.

    In case your wondering, I use Delphi and Borland C++ builder on Win. Starting to use Python also.
  • The good thing about all this, is we all have a choice.

    Personally, I am not anti MS or WinNT. The other good thing is bringing the shortcomings of both OS's to light. The only result(hopefully) is that both sides respond to critisism by making their respective OS's even better. I get the feeling that's what is happening with Linux, not so sure about MS.

    Linux, because it is free or low cost can answer critisisms with "okay, we'll fix it". At the moment Linux doesn't have to justify a corporation spending hundreds of thousands of dollars to license their software.

    MS would get into deep trouble admiting they needed to fix things, AND charge you lots of money while they're doing it. So, instead of honest and open discussions about WinNT shortcomings, they use FUD to preserve their revenue streams. Remember (if Alex St. John is to be believed) MS is no longer run by Techies, but by "professional" managers trying to increase the value of their stock options.

  • by shadrack ( 49555 ) on Thursday May 13, 1999 @06:32AM (#1894364)
    Some food for thought.

    1.A lot of sites use IIS 3, not 4 because 4 is still very buggy.

    2. Just for the sake of argument, lets say NT/IIS4 is better at the upper end. Is it $1,000.00 per server better?

    3. Which server has more server side application options? Apache with it's module interface is superior to IIS and ISAPI. For example, rewrite and perl or python are much more powerful than
    ASP and jscript/Visual Basic.

    4. Don't forget that to actually develop any thing worthwhile on NT/IIS (like database access and ecommerce) you have to puchase additional and sometimes very expensive tools. Tools that are free or low cost for Linux.

    Considering all the stuff you have to go through to setup any good website, the real answer is stability and options, not propaganda.

    Unless MS allows a truly disinterested third party to oversee and conduct the benchmarks, they don't deserve to have their stats believed, even if they are true.

    Just my .02

    Best Line not in the movie Matrix:
    When you can snatch the floopy disk from my hand Grasshopper, it will be time for you to leave.

  • M$ has an advantage here that linux can never overcome. CIFS is anything but common. The "standard" is a Microsoft brainchild. Running Samba on linux is an unfair test. The semantics of SMB match those of a windows environment, and do not align at all with UNIX filesystems.

    I would like to see NFS performance benchmarked. As much as can be said, NFS is the native network file system for Linux. I'd like to see how NT stacks up, regardless of hardware, when being pounded on by a network of Linux boxen.

    I would like to see Linux based web clients. Maybe a million win95 machines can open a ton of connections to a web server. But what about a few clients with highly optimized stacks that can saturate an OC-12 or GigE on the server?

    Maybe Linux does not play as well in the windows world as windows does. (I actually think it will win... How embarrasing, to be beaten at your own game!) But it is king in its own domain, where NT can barely operate.

    I work for UUNET Infosec. There is another concern here. My Linux based web server will still be here tomorrow, with MY content on it. And the day after, regardless of who does what to it.

    NT, IIS, and CIFS are a block of swiss cheese. There are so many holes, so many ways in.

    Perhaps they should run a set of tests against a web server while it is under DoS (denial of service) attack, run some real world traffic across the link while the test is running. Benchmarks are pretty useless if the machine does not survive the test....
  • by ebrandwi ( 49752 ) on Thursday May 13, 1999 @06:14AM (#1894366)
    Sure, NT may win. But if it is going to, the playing field had better be level.

    I started with Linux at 0.97pl3. I had to re-compile the kernel to change the IRQ for my bus mouse (37 mins on my 486). Things have gotten better since then. I remember when my roomate got the first Pentium in the dorm. PCI support for linux was miserable. If you FTPd to the machine, it would boot.

    Linux lags. Without massive commercial support, new technology cannot have stable efficient linux support immediately. Granted.

    But we now have 1 data point. With a machine above Linux's memory limit, with 4 CPUS (and linux SMP support lagging), and a poorly tuned web server, running a filesystem designed for another OS, NT wins.

    Now lets see a few more points on the graph. Tune the web server. Do NFS testing. Try a machine that linux excels on. I would not be surprised to find out that NT _is_ faster on the big machine. But maybe nowhere else.

    And then try that quad with gigs of RAM under Solaris. That'll give them an idea what Linux will be in the near future.
  • by SubDude ( 49782 ) on Thursday May 13, 1999 @06:00AM (#1894375)
    Hi Everyone:

    I skimmed the Mindcraft report and the first thing I saw was a highly optimized hardware package specifically selected to favour Microsoft.

    I have run the identical Microsoft software package on a client's P200 with 64MBytes of memory and an 6.4 Gigabyte IDE HD and it was a pig. The software was replaced by Slackware 3.6 with a 2.0.35 kernel and Apache and it runs
    like a champ. That box NEVER sees the kind of throughput that is contemplated by the Mindcraft/Microsoft test but most applications don't.

    If you can consider that most small commercial webhosting sites won't be seeing 1000 hits an hour, let alone ~3800 request per second.

    I have run into overworked and underpowered NT4 sites and it is pathetic because they are so ungraceful under pressure.

    Sure, it is easy with an unlimited hardware budget to get NT4 to run but that is not the real world!

    The real world is $5,000-$10,000
    hardware/software/consultant budgets running on 10Mb/s Ethernet connections. Linux and Apache are about the real world, not the fantasy world concocted by Mindcraft/Microsoft and their team of engineers working behind closed doors.

    I suggest to you that with hardware/software budget Mindcraft/Microsoft used and a free hand given to Linux/Apache techs would result in a faster, more reliable, more graceful solution (four distinct servers perhaps).

    Here is my challenge Microsoft/Mindcraft:

    Who can build the best HTTP server (hardware & software) with a real budget of $5,000US hardware/software/consultant fees total!

    Just one guy's opinion.

  • by Multispin ( 49784 ) on Thursday May 13, 1999 @05:59AM (#1894376)
    Did any one notice on the last row of the table they state that:
    PC Magazine (e[commerce - SSL) - 250 Requests/sec (680% faster than Linux)
    But on the linux side:
    PC Magazine (e-commerce - SSL) - 1950 Requests/sec
    Wait a sec here?????????????
  • Some points they made and my responses to them (MS: for things in the windows column, Linux: for stuff in that column):

    MS: "GUI-based tools" It seems having a powerful,flexible command line interface is a weakness. Hmm.. That's news to me.

    and along the same line:
    MS: "Wizards to simplify complicated tasks" and complicate simple tasks...

    MS: Binary backwards compatability, but at what cost? Some would consider that a weakness in Windows, as it requires large quantities of legacy 16bit code.

    another similarity:
    MS: "Extensive internal and external beta testing to ensure binary compatibility across services and applications" Except when you change memory allocation routines between SP3 and SP4, causing many programs to cease to function... (This should also be in the "binary compatability" section)

    MS: 99.9% uptime guarantee. Lets see, how many Linux boxen have I seen stay up for months/years on end, only requiring a reboot because of power failure or hardware/kernel upgrades? Even on cheap 486 parts. Sounds reliable to me...

    MS: Support for the latest hardware. It helps if hardware vendors like you and write drivers for you or at least give you the technical info you need to write them yourself...

    MS: "160k MSCE's". Um, OK. I work with a guy that is Microsoft Certified and let me just say that if he can get it, anybody can.

    MS: Scriptable administration on NT? And Linux has no scripting capabilities? I thought we had sh/csh/ksh, perl, python, tcl/tk, and numerous other script languages that can be used to administer Linux boxen...

    MS: "OS services and applications designed to integrated and work together" besides the incorrect grammar, it helps if you write the OS and distribute large quantities of software for it yourself. Nobody can do a better job of integrating with your software than you can.

    MS: "Clear longterm roadmap based on a customer focused vision". My vision isn't focused on having to buy a new computer when Windows 2000 comes out because my brand new 333mhz box isn't fast enough...

    Linux:"No application framework for developing distributed or Web-based applications" Hmm, Apache, PHP3, PERL, python. Seems you could make quite a few nice web based applications from those...

    MS: "End users forced to integrate (i.e., Web server, database, application authentication)" I look at it more like, "Linux users NOT forced to use the entire pre-packaged 'world domination' software kit. I run Apache because I chose it, not because it came with my operating system...

    Took a while to write this, so if someone else said the same thing while I typed, oh well... =)

Real Programmers don't eat quiche. They eat Twinkies and Szechwan food.