Slashdot Mirror
Red Hat/Apache Slower Than Windows Server 2003?
phantomfive writes "In a recent test by a company called Veritest, Windows 2003 web server performs up to 300% higher throughput than Red Hat Linux running with Apache. Veritest used webbench to do there testing. Since the test was commisioned by Microsoft, is this just more FUD from a company with a long history? Or are the results valid this time? The study can be found here."
Looking at the first page of the benchmark report, I see that they're using the exact same setup as in their highly contested samba benchmark, with a specific ancient version of Red Hat running on a specific hardware setup that version is known to have performance problems on. They could have at least tried a different server last time, or a modern version of Linux. Under fairer circumstances, who knows, IIS might have still won, but this rigged benchmark has nothing to offer us in deciding which server is faster.
10%? 15%? Those are numbers I'd believe. But THREE HUNDRED PERCENT? I like Microsoft, and I like when somebody defends them. But this is just bull.
Out of the box Apache doesn't do too well. But take some time tuning it, and your OS's TCP/IP stack, and you can easily outperform even Zeus. Read some of the tuning guides.
Questions: 1) Is there any reason to believe the results from this company would be valid "this time", if they never were before? 2) Can we please refrain from common basic grammar mistakes on the front page? 'There' is not a posessive...
Let's see. A test commissioned by Microsoft says IIS is faster than Apache. The link for more information goes to microsoft.com. Is this really "news"? Seems more like a thinly-disguised press release...
Simpli - Your source for San Jose dedicated servers and colocation!
kekelar muthafuckas. windows rules. linux sucks my pink cock.
Veritest used webbench to do their testing.
AKA a nazi fanatic loser.
....
...etc.
1. You rejuvenate and dance when you hear a windows flaw exposed, but you conveniently ignore the thousands of security flaws exposed in linux.
2. You yell loudly TROLL! at any person's post or at any person you see posting facts that you do not want to hear about your oh so cool linux.
3. You know it's a classic case of penis envy, you don't have all the support, software and hardware available for linux and you have to let that anger out somewhere, but you don't have the brains to admit it.
4. You hate windows, hate Microsoft, but race to emulate windows, have programs to run office from within linux, and spend a $300 on a Windows emulator, only Windows fools.
5. You cannot admit that you don't have professional usage of Linux outside server markets.
6. You cannot admit that most of the joe user out there when told that there is linux will respond, what is that?
7. You cannot admit that there is no professional printing capabilities in linux.
8. You cannot admit that you are a masochist (otherwise why would someone spend hours playing with scripts,
and recompiling programs that are available for Windows?)
9. You cannot admit that there is no professional desktop publishing done on Linux.
10. You cannot admit that no one in their right mind would do professional video editing in Linux.
11. You cannot admit that linux sucks when it comes for gaming/home entertainment or education.
12. You have problems in understanding Windows, and you will blame your own incompetence on Microsoft.
13. You have problems in pointing a clicking, but have no problems in wading through cryptic scripts written by lunatics.
14. Nothing will get past that shit that fills your head, you will not admit to any facts.
15. You can't admit that naming of linux components, packages, and others are weird and fits profiles of troubled teenagers. gentoo, lgx, rpm
16. You feel angered because you were left out by microsoft's Media technologies, they support Mac, Sun sparc, but not linux.
17. You feel inferior deep inside but unable to admit it, you don't have a database as easy and powerful as Access.
18. You cannot tell that not a single office package outside Microsoft's is worth looking at or bothering with.
19. You don't know that your CD recorder software sucks.
20. You don't have DVD-RAM, DVD-R, DVD-RW support in your pathetic OS.
21. While the rest of the world moves on, you're stuck in a stone age technology that needs third party software to boot into GUI.
22. You act out of prejudice, you kill file domains and users of specific news readers while you ignore the bullshit that your fellow linux losers post.
23. You don't know commercial support in Linux is almost non existent.
24. You miss the fact that companies are leaving linux because of the chaos, and the cheap linux losers who are unwilling to pay and support hard work, Corel, gaming companies,...etc.
25. You are unaware that linux has no terminal services (there is a lame one that no one uses), and commercial support for it is not happening.
26. You are unaware that setting up servers on Windows takes couple of minutes while on linux, good luck playing with configuration scripts.
27. You cannot admit that support for USB on linux is laughable at best.
28. You think that Linux is better because slashdot told you so.
29. You spend countless hours flaming people because they post their opinions about your oh so cool linux and your attitude, instead of researching things for yourself and understanding fact in order not to look this stupid.
30. You think that anyone who uses linux has a clue.
31. You think that linux cannot crash.
32. You think that everyone is interested in your conspiracy theories about Microsoft (or should i say M$ in order for you, teenagers to understand?), and how they destroyed linux,
33. You keep ignoring the fact that thousands of linux servers get hacked every year, but it takes one Windows server hacked to get you and your fellow linux idiots to dance and celebrate.
At least they're up-front about it these days.
Other Veritest-Microsoft fun:
http://www.veritest.com/clients/reports/microsoft
http://www.microsoft.com/windowsserversystem/fact
http://www.gotdotnet.com/team/compare/veritest.as
In short, this is a company paid by Microsoft to make reports/whitepapers that make Microsoft look good. Nothing wrong with that as long as everyone's aware
rooooar
"Since the test was commisioned by Microsoft, is this just more FUD from a company with a long history?" Did you expect anything less from our global propaganda machine?
Microsoft Sucks, F/OSS Rocks. I get mod points now right?
Notice the total lack of the CGI script?
And the results are interesting. The Gentoo server doesn't perform nearly as fast as the Windows Server for most basic serving tasks. But software like Exchange Server is so badly written that it's much slower than postfix.
It's sad. If the same people writing 2k3 were writing products like Exchange, we wouldn't have a need for the Linux server.
Faster to get infected.
Faster to get rooted.
Faster to get used as a warez server.
Nothing new here.
Reminds me of this editorial on the G5's testing by Veritest. http://spl.haxial.net/apple-powermac-G5/
So does that make SMS on Windows faster than morse code on Linux?
I wonder if Bill Gates actually believes his own bullshit...
Microsoft Sucks, F/OSS Rocks. I get mod points now right?
If they were running heavily restricted SELinux on RedHat it wouldn't be surprising to witness a massive slowdown on certain applications, and will likely be infinitely more secure than a Windows box probably could ever be. Beyond that Apache can be very slow out of the box, on my hardened gentoo test system (please withhold funroll loops jokes) Apache2 with hardened PHP + MySQL I would be lucky to handle 2 requests a second happily, it was amazingly slow. I've yet to fully tune it but some even basic tuning was able to improve speeds dramatically. It wouldn't surprise me if similar techniques were used by this "benchmark".
What possibly possessed them to publish these results. No one in their right mind is going to believe 300% is an accurate figure under fair testing conditions.
Philosophy.
PS: Darnut, are these Veritest guys going to get a better team99 bonus than me?
Did anyone else notice the versions of RH they were using? So does this mean that Win2k3 is equiv to RH AS2.1 and IIS6 is equiv to Apache 1.3.23.
no sig yet
how many times are we going to hear about how cheaper/faster/better windows is and from how many different companies? no, i'm not implying they're right, and even if they were, is it news everytime it's claimed?
"Veritest used webbench to do there testing."
Disgusting.
It might just be me that is a nitpicky person, but there is a difference between "there," "their," and "they're." This is not only simple grammar, it's basic grammar.
Why couldn't IIS be faster than Apache?
Is Apache/Linux the "end-all-be-all, there is nothing that can be better so let's stop trying" type of quality?
Are the guys who work at Microsoft a bunch of idiots that anyone can out-program?
I'm sure IIS is better at some things, maybe more things, maybe less.
Who cares! I don't think stats like these are why anyone chooses Apache/Linux over IIS/Windows.
There's a reason why it's not legal in most of the civilized world for witnesses to be paid to testistify.
I've never been a Microsoft fan but if they would pay me the kind of money this company received, I'd probably say whatever they fucking wanted me to say.
Hey slashtwats...when the 'study' is found at MIT or Berkely, or Waterloo or that would be unbiased (or has some semblance of credibility for that matter)...wake me up, m'kay?
This is why really high traffic websites run simple httpds like thttpd which is very small and very efficent, unlike Apache.
In TFA: Published: May 5, 2004
Who wants to bet this is a year old dupe?
"Orthodoxy means not thinking--not needing to think. Orthodoxy is unconsciousness." --Eric Blair
... do not welcome our new Micro$oft overlords.
I'll test the amazing Linux versus the ultra-slow windows NT.
Config:
Linux: Latest Redhat running on Opteron 4GHz
Windows: Windows 3.1 running on a Pentium 100.
And the winner is...?[/sarcasm]
Windows 2003 server running on skynet is 300% Faster than Ye-oldie redhat -12 edition from 1723 running on an abacus.
This reliable Expensive test paid for by Microsoft to show how much better windows 2003 server is(the payment came with a clause stating such).
The only things certain in war are Propaganda and Death. You can never be sure which is which though
Right at the top of the PDF it says "April 2003". How is this benchmark "news"? (And nevermind the fact that as always, as an MS sponsered benchmark, the MS machine was probably hand-tuned, and RH + Apache was probably run in a stock configuration.)
While sheer performance isn't really what sells RHEL boxes, I'd be very interested to see a proper test of Win2k3 vs RHEL 4 on identical hardware...
The Free desktop that Just Works
And remember, that the TC0 (0 for 0wnersh1p) is lower for Windows as well (""Immunity's findings clearly show that the best platform for your targets to be running is Microsoft Windows, allowing YOU unparalleled value for THEIR dollar."). For anyone who missed it, /. had a lot of great discussion on that one from people who couldn't detect a troll.
The web page says it was published May 5, 2004, i.e. a year ago. The report itself is dated from April 2003. The test was done using RH advanced server 2 and Windows 2003 RC2, i.e. a pre-release version. Since then, both RH and Microsoft have published new releases, for example the service pack 1 of Windows 2003. Why is this posted now?
Not only do we get MS ads at the top claiming the same stuff as the article...now we get articles promoting it.
If it helps keep slashdot online...fine...but this better be a rare thing.
Sheesh -- with such outdated news, I almost felt like I was reading the newspaper or something.
...but what about reliability? Is a Windows Server reliable enough? Can I sleep peacefully if my server is running Windows? Can it give me uptime of 364/24? I bet Microsoft won't be answering these questions.
their
HE SPEAKS THE TRUTH!
Here's an article from linuxworld. Dated but shows that Veritest makes mistakes in setup.
Google should switch then.
Wonder what that benchmark would be if you installed the FULL Norton package on it?
This bull reminds me of those advertisements for weight loss.
BEFORE................AFTER
Stick stomach out....Suck stomach in
White......................Tanned
No cosmetics..........New facial
Front shot...............Side shot
Grubby clothes........New fashions
The parent is not a troll, thttpd is actually a faster httpd than Apache while running Linux.
Please fix, mods.
You MS dorks set up your super-Solitaire computer, and let me and some Linux-heads set up the linux computer, and submit them for competition.
2003 has kernel-level webserver acceleration and offloads a lot of the processing
there, the same was as the Tux webserver (also RedHat?) beat the shit out of
Apache. It's essentially zero-copy-networking with zero-copy-webserving too.
http://www1.us.dell.com/content/topics/global.asp
There may be some truth in it, therefore. Aren't there some patches these days to
hook Apache directly into the Linux kernel too, since Tux is obselete? I doubt
they ship with RedHat's stock system though even if they exist.
I remember back in the day a 486 with enough ram could fill a T1 pipe with apache, while the windows guy would still be trying to coax the installer to run(being well below the minimum requirements). Now it's never that simply, static HTML is a dinosaur that doesn't exist, and how many pages do you visit regularly that are static html? There are some out there, but which are regular(daily/weekly) visits? Yeah, there are the few exceptions, but 99% of what you see needs a CPU and Ram. Windows could have a chance here, but I'll stick with my PHP/Apache/MySQL with eAccelerator.
It's a different world and benchmarks like this just don't give a good picture, not yet anyways.
I find it odd that if Windows is so fast that the performance of Hotmail is so atrocious when compared to Gmail or Yahoo which both run off of Unix based systems.
I mean Hotmail page loads often quit right in the middle!
Riddle me that one.
Knowing MS they probably optimized the OS for the benchtest in favor of providing their customers some real world performance.
Replying to my own post, sad...
Does anyone know if any of the bits of the Silicon Graphics accelerating apache project were ever rolled into Apache 1.3.x or 2.x?
http://aap.sourceforge.net/
KDE and Gnome are laughable at best
I thought it was a little odd that the test setup included the HTTP 1.0 protocol. Without 1.1, you don't get the benefit of persistent connections. While both 2k3 and Apache could take advantage of these persistent connections, I thought I remember a rumour that IIS used (basically) hacked persistent connections between IE clients and an IIS webserver. Therefore, running HTTP 1.0 limited Apache's ability to limit the overhead of setting up and tearing down tcp/http connections repeatedly. However, 2k3/IIS limited this overhead by using techniques not allowed by the official protocol. It seems like Apache is being punished for following the rules. Just one opinion.
remove nospam. to email!
Different environments are good at different things. Now I'm no expert in web servers, but I don't have a hard time believing that you could find a scenario where IIS stomps Apache. If I were going to look, it would be a scenario that used some kind of scripts with lots of DB access. IIS allegedly features wonderfully fast intergration with .NET and MSSQL. I know that with IIS 6 .NET code can run in the server process, eliminating a context switch. Thus I find it perfectly feasable that one could find a scenario where IIS is much faster.
The real question is how relivant to the real world is it, and under what circumstances. It is, perhaps, something that is rarely, if ever, done in the real world. Or if used, something such that the expense of server hardware is trivial to other costs and doesn't really matter.
However to simply dismiss it becayse you feel that it's too big a difference is silly. That kind of thing happens in real life. A personal anecdote:
My roomate, several years ago, was hired to do web work for a small startup. Their immediate problem was their website was sloooooow. Client querys took forever. He took the first logical step and upgraded their very underpowred hardware to something top of the line. While this helped, the system still lagged, and avereged about 10 load. He then turned to the software, a PERL script accessing a Postgres DB. There were further problems with data loss, etc.
Finally, one weekend he got fed up and recoded teh whole thing to move over to MySQL and ModPERL and brought it back up. The load average then dropped to about 0.01. That's a speedup of 4 orders of magnitude.
Why is this news now? Wasn't Linux still on the 2.4 kernel then?
Don't see how you can call this "News" for nerds or anybody else. The editors are asleep at the wheel again.
I've been around on the net for a while now and if there is one thing I can say that is universal it's that servers that implement ASP are generally more flaky than other types of servers.
I use tvlistings2.zap2it.com which has ASP, and while I think they've gotten far better in the recent past, even 4 or 5 months ago, it would routinely lose my channel line up and if I'd try to log in to reset the cookie it would claim my login account doesn't exist. I'd follow their suggestion and try recreating the account and it said it was already in use. But I can't log in because it doesn't exist, but I can't recreate it because it already exists, but I can't log in because it doesn't exist.......
Anyway, I notice time and time again how sites that churn out ASP pages have typically slower response times compared to ones that have PHP or straight static HTML. For anyone who wonders how I determine that, I go to load a web page, and I wait for it to load. If it starts taking a while and I mean a really long while, I look at the URL and more often than not, I'll see it has a reference to an ASP. Maybe the "oh it's another one of those stupid IIS servers" makes it stick out in my mind more than "wait, this one is slow. I don't really know what's running it but it's crap", but if I had to put money on it, I'd say the IIS servers are generally slower.
I don't run a web server, I could, but I don't. Managing web servers would not be a job I'd want to do. Almost all of my web server experience is on the visitor side and without any kind of overtly blatant bias from any sources (like the kind of "windows crashes therefore windows is evil and anything dealing with windows is also evil") to affect my opinion, I'd have to say that I personally experience a more significant lack of performance and reliability visiting web sites that run IIS than other sites that don't appear to run it. So to me, a report like this is microsoft's ever so polite way of trying to stick an uncomfortably large tube up my ass and then proceeding to blow smoke through the opening.
Anyone do the math to see what that would cost.
It's conventional wisdom that Google has about 100,000 servers. If google went with Windows 2003 Server Enterprise Edition (which costs $3999 ) That would cost google about half a billion dollars.
Extending the logic to use SQL Server Enterprise Edition as their search database, at $25000/server the price would go up to about $2.5 Billion.
Every CEO likes to be like Google and likes talking about numbers like billions of dollars; so this is a fun set of numbers to throw around when your're discussing microsoft partnerships with the CEO.
(Note, however, that in the true spirit of Team99, I must say that Longhorn will make it well worth the price, though, and I wouldn't be surprised to see Google switch)
...it's the license, stupid.
pr0n - keeping monitor glass spotless since 1981.
People keep saying, 'When are we going to get a real benchmark?" Well, why don't we roll our own? Seriously.
Here's my idea:
Slashdot has strong zealot^H^H^H^H^H^Hsupporters for both Microsoft and Linux. Let's have a contest to select the best qualified from each side, have them work in teams on identical hardware. Let them make any changes, tweaks or optimisations they can dream up. Then, let 'em rip.
I'm dead serious about this, by the way. Let's get off this endless roundabout and for once make a clear comparison.
For bonus points, once the first contest is finished, we should take the two servers, leave them exposed to the Internet and see which one gets 0wned first. 8^)
Crumb's Corollary: Never bring a knife to a bun fight.
From the original Microsoft link, you can see that the release date for this April 2003 study was 5 May 2004.
My guess is that the Slashdot editors confused this with 5 May 2005, which was yesterday.
Don't try to use the force. Do or do not, there is no try.
Microsoft argue that Apache is slower because CGI is slower. They say that it needs to spawn a new process for each request, which is correct.
But how many years have mod_perl and mod_php been around now? Does anyone actually use CGI on Apache this decade?
Perhaps a more fair comparison would have compared CGI on IIS with CGI on Apache. And I'm pretty sure that for various reasons (spawning processes is slower on Win32 than on Linux) IIS would lose horribly.
Karma: It's all a bunch of tree-huggin' hippy crap!
This 'study' has been around a while. Here's the premise, take the latest greatest windows bits and compare them to 4-5 yr old linux stuff. RH 2.1, 2.4 kernel and Apache 1.3. When this study was done, the 2.6 kernel and Apache 2.0 were well established.
Suppose you were an idiot. And suppose you were a member of congress. But then I repeat myself. -- Mark Twain
Actually, it's learned behavior. We've seen so many fact-warping MS-sponsored studies, astroturfing campaigns, dissembling regarding the nature of their monopoly, and other aggressive PR that it's no wonder people are more than a little skeptical.
This reminds me of something someone told me about graphic card benchmarks. He is a 3d graphics professional, and he was called in by a rather large chip company to help them in a test against another large chip maker's video card. The arrangement was that he would work with the representative from the other company to come up with a "fair" set tests to which both sides could agree.
As the more experienced guy, he was able to get his counterpart to agree to tests that worked squarely in favor of his company's card. This is in a scenario where it is supposed to be evenhanded, since both companies agreed to the test methodology.
So it's bad enough already. Compare a situation like that to one in which Microsoft is commissioning a study, and you can imagine why people react with such profound skepticism.
Read the EFF's Fair Use FAQ
I actually thought that this was common knowledge--that Windows Server 2003 with ASPX was faster than Linux/Apache with PHP, or that Server 2003 was generally faster with static content. (I admit, I only glanced over the article, and Adobe Acrobat's search tool is the worst of crap, so sue me if it didn't mention ASPX).
.NET framework for web content delivery (get it--the 'X' makes it sound cool. Or something). .NET is compiled, and ASPX needs neither process nor thread creation. Like any .NET application, ASPX can run sort of close to native speeds (native + lots of wrapper overhead + generic memory management overhead and such.)
1)ASP (not ASPX) are fairly flaky and recent versions are roughly comparable to, but slower than, PHP4 (not sure about 5), in general.
2) Windows is not very good at creating new processes quickly. This is why CGI (not fastCGI) in the platform is so glacially slow.
Let's have an example. Let's say that you make a dynamic webpage in which all content is generated by a C++ CGI program. Ignoring database access for the time being, since that dilutes the example, on Windows, the website would be MUCH slower than the same website written in ASPX, even though the actual execution time of the C++ program is shorter (assuming a competent C++ coder).
This is because for each request, Windows must create a new process (the CGI program), and destroy the process when the request is complete.
While the execution time is low, the process management overhead dwarfs the actual page runtime, because Windows doesn't do that sort of thing quickly. This is why CGI has long been blacklistedon Windows systems by good web devs, and this is one reason that Apache 1.x was such a dog on Windows. Apache 1.x creates a new Apache process for each request.
Now Linux, on the other hand, creates processes about as fast as it creates threads, which is to say, really damn fast. Apache 1 has always worked just fine on Linux (and indeed most Unix systems) because the overhead of creating a process, while significant, isn't slower than a dead slug stuck in frozen molasses like it is on Windows.
Apache 2.x allows requests to be served by a thread or a process, or a number of processes that each create several threads (any Apache gurus please correct me if any of this is off).
It follows that this isn't a big deal on Linux (because process creation isn't really much slower than thread creation), but is a very big deal on Windows.
Windows has ASPX, which is Microsoft's marketing term for the use of the
Yet Apache is still back here creating a process or thread for each and every request (note that there are some ways to speed things up. FastCGI comes to mind, but I don't want to get into the gory details that I don't know enough about). This is not the brightest way to do it in terms of performance, but then, Apache appears to have been designed for universality and configurability over raw throughput.
It is unwise to hold the attitude that Apache can't be beaten by IIS, especially when IIS is optimized for one platform--by the vendor of that platform. Apache isn't even the fastest on Linux. Take a look at Zeus webserver. It serves circles around Apache on any platform it supports--including Penguin land.
In fact, Zeus uses a technique called SendFile() which, oddly enough, is strikingly similar Microsoft's own TransmitFile() API. Hmm.
Think of it this way: Apache is to IIS as GCC is to ICC, at least in terms of performance and generality.
Intel's compiler (ICC) consistantly blows away GCC in terms of the performance and size of the compiled code, but GCC runs on just about anything with a CPU, can cross-compile, is free, doesn't pull any PHB evil tricks, and actually compiles things like the Linux kernel without pat
Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
Just look at the cover page. The report is dated April 2003. Why is this news once again please?
the overhead from running 3 OSs combined would slow things up too much. I guess Win98 crashing early on would help though....
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
Veritest or Verisign? Which one pisses me off more.
AAAH! I just say an MS "Get the Facts" advertisement on the top of slashdot!??
supposedly...
"Windows was far more cost-effective than Red Hat Linux." -J.E. Henrey, CIO, Regal Entertainment Group
OK. Was. And why would I take the word of a CIO of some company that isn't even in IT?
Will the 70% (happy) Apache users switch to IIS+Windows2003 after reading such a lame old piece of junk?
If you don't fail at least 90 percent of the time, you're not aiming high enough. (Alan Kay)
If they don't run those tests with antivirus protection on in the Wintel server, it is not anywhere close to the real use case performance.
If someone publishes a benchmark about your software, and finds out your software does not perform well, don't whine, don't behave like a child, don't start kicking and screaming, don't tear his hair out. Behave professionally.
Good starting points:
Let me summarize what I think about their test. First of all, I believe their numbers. Apache sucks performance-wise, in particular if you run a busy site with dynamic content. That's why people are using squid in local accelerator mode before Apache. This is a good indication that some performance tuning is in order. But no, people rather wait for Microsoft to find out and then they start thinking about fixing it.
If this test was meant to be unfair FUD, they would not have tested TUX, just Apache.
But now to my questions above:
Question 1: is their setup relevant?
No. Sites who answer more than 5000 requests per second are not using a single web server, they are using a load balancer and a cluster.
Question 2: Can their numbers possibly be true?
The point I find least believable is that IIS had better CGI performance than Apache. Creating a process is really slow on Windows. Their result should be independently verified.
Question 3: What weak spots about the competition does their test reveal?
They did not test a single-CPU webserver (which is what almost everyone is using).
They did not test FastCGI or APAPI dynamic web pages.
So if we wanted to do a more balanced review, we would look at these.
Question 4: What can we do to improve the results.
Document APAPI better, I'd say. Almost nobody is writing their dynamic web page modules with APAPI.
Everyone is using PHP or mod_perl. Benchmark Apache in real-world scenarios. Document best practices.
Thats all there really is to say. Skewed, yes.
Quack, quack.
Allowed HTML: ... <a> ...
Can anyone tell me how do I use that?
<a href="...">...</a> does not work.
If you don't fail at least 90 percent of the time, you're not aiming high enough. (Alan Kay)
Spend $500 on each machine.
Instead of $30,000 dollar machines.
Remember this is FREE software they are testing against.
Wow microsoft says they can beat free software.
I don't really care if Microsoft has managed to change their settings around so that their server goes faster than apache on linux?
Microsoft through they are for the most part formidible businessmen, has lost my trust to such an extent that I don't give a damn about their products. In the news...the way Google is making revenue is sketchy, but I trust them so I use their services anyway.
Now lets look at this. If RedHat were to pay someone to test thier (the communities) software would you say: It was paid for by RedHat so the results can't be trusted?
To put this to rest RedHat and Microsoft should get together and hirer two companies to test thier software, both agreeing on what to use. I know this will never happen because BOTH companies are too fearful of the results.
I have been a Linux user for many years (since late '94) but I am getting tiered of all the extra configuration that I must do to get it working the way I want. I hate to say it but more and more I'm using windows.
I still use Linux and will continue but (and there is always a big BUT) Linux has its place.
There is one undeniable fact, Linux make Microsoft software better as Microsoft make Linux software better.
One last thing: Where are the tests compairing MacOS with Linux and Windows???
As MAC OS X USER (already you hate me) -where the hell do you get off talking about video editing, printing, desktop publishing and, OF ALL THINGS HOLY, UI (I think this covers the lot of it)?! Oh wait! The part about hacking is good too!
If I may be granted the graciousness of smoothly transitioning into the subject: The big three pro-sumer OSs (Windows, OS X, and Linux) form the symbiosis between business, creative, and cutting edge tech -respectively. Don't dis'!
Shizzz, 'talk about being a "looser"
Even if IIS were faster, it's a no brainer to use Apache just for its shere scalability and powerful features.
One would have to be an AMATEUR to go and run a proprietary IIS web server that will only run on Windows.
And, as long as I'm going to be implementing Apache, I'll be implementing it on Solaris SPARC/x86/x64, HP-UX and SGI IRIX, not some AMATEUR-LOSER OPERATING SYSTEM like Linux.
Called the FUD section with a nice new icon of Elmer .
having this artical under the linux section just dosn't sit right with me as its really about windows and MS bull-turd bias studys.
Hm a new Gates icon with an Elmer F.U.D hat on would work well
The only things certain in war are Propaganda and Death. You can never be sure which is which though
Yes, I agreee. What is the world coming to when people get a little upset over a few idiots who take the worlds greatest communications medium ever created by mankind and then totally fail to communicate properly with it?
Seriously, it's like using a wrench to drive a nail. It makes you a fucking idiot because you're abusing your tools.
I don't know if these numbers are trustworthy, but at least its another datapoint.
Maybe a real-world comparison would be more meaningful, like Slashware on Apache vs. IIS, with a real load.
Anyone who has actually used both knows that W2003 is light years ahead on anything else out there...so is this really news to anyone?
And with "anyone" I mean "anyone who actually knows anything", not the rdf -people or puta (pinguin up their ass) -people.
Oh well.
What is that one thing about the scientific process that is hammered into every k-12 students head? Reproducibility based on recorded procedures so that others can varify you results. So if you will are uncertain about this gem of marketing/benching reproduce and see what realy happens.
Time and time again slashdot reports about a Microsoft commissioned look into Redhat or Suse running Apache vs. Windows with IIS. These are a waste of people's time. Plain and simple. Speaking from an ISP/Webhost that switched from Win2k3 to Linux (notice no redhat and suse mentionting) and apache I get much higher performance from linux and apache than windows/IIS could ever provide.
/. wasting front page space for something like this is literally insulting to those of us (and I think you know who you are) that know the truth.
On that note,
I mean...IF, god forbid, windows with IIS was able to out perform linux/apache it's either because of....
A) poor configuration on the linux admin(s) part
b) using nothing BUT MS proprietary stuff that runs all through "integrated" programs and thus becomes kernel level which is unsafe as hell
c) Jaded results looking at ONLY what IIS does better than apache.
I'm tired of this "debate" and having seen both sides of it, I know the truth. Which is why I'll stick with my linux servers running apache....
There's a reason why I'd not use IIS (again). First of all, I really love the freedom of open source software. And to get back to the benchmark, well, I just think that the apache server could have had some better circumstances. To be honest, I think if I set up an apache webserver it would not run slower than the IIS (hey, it's just god damn paradox). But, apart from the apache-vs-iis-war, I can only recommend using http://www.lighttpd.net/. It's really fast, simple and secure, unlike others.
Given Microsofts record as a vendor supplying expensive low quility bug ridden crap, I find it hard to believe any survey which claims that their products are any good.
I don't believe I'm alone here.
How about next time we just skip the dated biased benchmarking report and post "Somewhere, at some point, there possibly might be some pro-Microsoft anti-linux FUD"?
The real problem with reports like this, is that they are directed at IT managers, not techies. Everybody with some webhosting knowledge sees through this bias. But most managers (like I have one at work... he used to work for MS) don't know shit about it, and probably never touched a webserver. But still those retards are the ones who got to make decisions.
Something else... why didn't they include a Win2003 running Apache in their test?
Red Hat and Apache both have been poor performers. There are other web servers out there if you prefer performance over support and modularity. And other distros if you prefer speed. although just building a new kernel instead of using the stock Red Hat kernels generally makes Red Hat as fast as any other Linux distro.
:)
2.6.11.8 has been out for 7 days, why aren't you running that? People might start thinking you're obsolete if you don't upgrade.
“Common sense is not so common.” — Voltaire
Or am I alone when I say I couldn't care less about yet another MS-founded Linux-bashing tests ? For some reason they always pop up on /.'s front. Maybe you like'em so much, well, I have nothing against that, it's your right to like whatever you want. I just feel these "news" are not worthy of more than a sidenote on some last page among the worst ads.
I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
I see a lot of complaints here, most of them valid. But why noone does anything about it? I mean do your own test on a hardware of your choice, send an invitation to microsoft to tune their OS, you tune yours, and both parties should choose half of the tests. And then publish the results. (GetTheRealFacts.com would be nice) I'd like to see them, probably others too. If you let microsoft publish only their goodfacts, sooner or later they will win non-professional (executive) minds, which is their goal, i suppose.
That's all I have to say...
Yeah, free Ipod! He is innocent!
Is it true? Does Microsoft's server products actually force you to run a graphical desktop environment? Is this desktop environment even more bloated than the standard Windows XP one with extra tools everywhere? Is it true you can't turn off the desktop environment and work through either a web based control panel or a command line? I'm just wondering. If a company does these things, it's pretty unbelievable that it'd run better than a real server.
http://illhostit.com/ - Webhosting
TUX is still being developed and it's shipped with even the latest version of Red Hat (no surprise if you consider that TUX started as Red Hat Content Accelerator).
It is installed by default if you select Web server as the installation package, but it is not started or configured. You have to tweak the things a bit to run it, but when it runs, it runs great, thank you very much, with TUX serving static pages, and using Apache for stuff it doesn't understand.
I'm only not sure how it is integrated with newer Red Hats (RHEL3U3 and RHEL4) since SELinux was introduced.
Just before clicking in the "here" link... I saw that the URL was from Microsoft... come one!, they surely won't say that the RedHat box was better!!
/real/. Can anyone point to a page with a benchmark done by a third party company??
Now, It would be interesting to see a benchmark commisioned by RedHat... would it be all the contrary?? (RedHat beats by 300% to IIS) or would it be more
Ubuntu is an African word meaning 'I can't configure Debian'
If Microsoft are so goddamn confident of their product, and so keen to rubbish the Linux community, they should challenge the Linux community to a server benchmark shootout.
Get equal representatives from each side to devise a set of tests, and agree on a standardised hardware specification. Then each side is allowed to hand pick the engineers that work on the builds of each OS and all the fine tuning and tweaks, and run it as some kind of well publicised event.
If Windows Server 2003 is so fucking good Microsoft would have nothing to lose.. the cost of marketing an event like that would be nothing compared to the good press and extra business it would give them. But quite honestly I feel Windows 2003 WOULD get screwed over by Linux and the negative publicity would sound the death knoll for Microsoft. And thats why they'd never do it.
Maybe the Linux community could formally offer such a challenge to Microsoft, and bloody publicise the fact that they are trying to do it.. if the press leapt on it, Microsoft would have to have some spectacular excuse to back out of such an event?
"Hey! Unless this is a nude love-in, get the hell off my property!!"
A team of "experts" with each OS to tweak identical machines in advance of the test. That way, it's more like F1 racing where the cars have more or less identical performance and you're seeing differences in the software (or driver) choices.
Yes? No?
Cheers,
Is anyone keeping an online List of these research projects that MS has fundes? Like teh Halloween Documents found at OSI
Or in other words has there been any honest research on MS research?
Its becomming Important, as this shows persistant intent to deceive the consumer. Isn't consumer deception Against the Law?
insert fark tag here
Veritest? Don't tell me, Kilroy Silk did the test?
Hmmm.. might explain it then!
Give both systems limited amount of time and money. X dollars to buy hardware (and maybe software). Y hours to set up and tune the systems. After that run the benchmarks. The system that wins has pretty much the best bang for the buck.
I mean, 200% to 300% !?
Can't they tell a lie? If IIS won by 20% to 30% less geeks would be outraged and they would stand a chace of not being exposed...
Its the biggest software company in the world we're talking about, they sure can hire the best team to produce a fake study!
I really hope they can learn from their mistakes, and can produce a better and more beliaveble faked study showing how Longhorn is more innovative than MacOSX Tiger.
---- You know how some doctors have the Messiah complex - they need to save the world? You've got the "Rubik's" complex
Zonk posts always crap.
If you're looking for a webserver and webserver only, this sort of comparison makes sense.
BTW, the german C't magazine did a test like this once, with IIS on the newly released Windows 2000 versus Apache on Linux with a beta release of Kernel 2.4.
They did quite a variety of tests, with Apache/Linux looking slightly better in the overall result. But the difference in performance was not large, so other considerations like price, support and security should have decided your choice if you wanted a web server back then.
C - the footgun of programming languages
Published: May 5, 2004
Correct me if I'm wrong but I think the majority of us have moved into 05. This article is part of the aborted Get The (MSFT Financed) Facts.
Either that of I've suddenly been transported into last year, sort of longer version of Ground Hog Day. That could be kind of cool. I can keep living 2004 over and over.
If this is really 2004 again, then I'm not going to spoil the surprise of what happens in 05.
That's our life, the big wheel of shit. - The Fat Man, Blue Tango Salvage
you have to go back to 2nd quarter 2003 to find a hardware vendor using IIS. Intrestingly, that vendor was Dell showing off a PowerEdge 2650 running IIS5 as well as Rehat Linux 8 (link http://www.spec.org/web99/results/res2003q2/). On this hardware, IIS perfomed 3% better than RHCA (TUX) on Redhat Linux. So, if speed is what you want, hardwave vendors have given up on IIS about two years ago. If features is what you need, then it seems Apache is the choice. Seems to me, IIS has no role to play any more. There's always a better and cheaper choise elsewhere.
Showing how much better Windows is than Linux it would be interesting to have the following stats to make sense of it all:
1) How many surveys did Microsoft commission on this subject?
2) What percentage went through the shredder because they didn't say what MS wanted?
Microsoft sees a marketplace with Linux in it as a battlefield and as the sayings go, all's fair in Love and war and the first casualty of war is the truth.
Hmmmmmm..... Deep fried and look like Squirrel.
The red flag here is "300%." I don't think anyone can take it seriously with such a large desparity. That's like two hybrid car makers competing and the salesman tells a customer "Yeah, they get 50mpg but ours gets a bagillion-zillion!"
They tested it agains redhat 8! (yes, redhat 8 was one of their test cases)
My new blog
Since its MS commissioned. What do you expect? A fair comparison? MS, fair? HAHAHA!
Seems like Bill is everywhere lately. From bashing Linux, spreading FUD, bashing Nintendo...
I know this is off topic and im going to get shot down like a kite over Bagdad, but the Windows GUI/environment is faster than many X setups ive seen, im not saying its Linux, im not saying its X, im not even saying its KDE (ok maybe i am), but my point remains valid, Windows might be expensive, crash-prone, bug-ridden, lacking of features and full of security holes, but it can still hold its own on speed in some arenas and im just saying _maybe_ its possible that Apache or Linux or someone in that chain is not pulling their weight.
This comment does not represent the views or opinions of the user.
Don't bother about this 'test'. The webpage at MS is one year old, the study more than two years. If MS IIS would have beaten Apache 2 on Redhat Enterprise Linux 3 by now, they would for certain have published those results with relish. They didn't. Instead those marketing people at MS just let the same nonsense stand without apparently being able to do better. I feel rather sorry for them...
The article linked to states "Published: May 5, 2004".
In case you hadn't noticed, it is now 2005. Not 2004. What kind of moron editor doesn't even catch these basic errors?
From now on, let's insist that they connect the Linux server to the network while they are running the benchmark.
My company has purchased testing through Veritest, and I can tell you they suck to an extortionate degree. Read on.
They were hired to measure web server performance using our hardware. Unfortunately for us, the only testing tool they can use is WebBench which contains serious bugs that prevent it from measuring full speed correctly. I only know this because I experienced the same problems in my lab as they eventually blamed on our product. When we developed our own test tools, we discovered it's not that easy to make the benchmarking code perform well enough to handle the load our product carries. But we had to figure it out in order to push full to our product and we know it's robust because of that.
We tried to take it up as a defect in their WebBench but they were unwilling to consider their own fault. They wanted to charge us $engineering time even if the fault turned out to be their own.
Lessons learned: 1, caveat emptor. 2, Veritest sucks.
If you require identical hardware, some might complain that it's more suited to a particular system. Give them a fixed amount of money for the server. Or a fixed amount of other resources that might be the bottleneck (power, floorspace, maintenance time/month,...)
What keeps me going is my inertia.
Not worth squart when it comes to anything objective.
I'm not a troll, but I play one on Slashdot.
1.you make stupied lists against linux on forums
2.none of your points are true
On a long enough timeline, the survival rate for everyone drops to zero.
What's that author been smokin?
When have you seen a study funded by Microsoft that didn't show that Microsoft was better?
All i have to say is.. rock the fuck on. Thank you for a funny/good read this morning (it's 8am on sat. morning).
Of course, the budget has to buy both software and hardware.
:-)
After all, we want to count TCO and performance, right?
Good LORD, this should have been the first reply to the article. I can't believe how STUPID these geeks can be sometimes - get a BRAIN, MORANS.
Holy Christ, how did you guys NOT pick this up.
If IIS is really 4x faster then Apache. In most cases it really doesn't matter. Rarily people who are connecting to your side across the internet will reach the bottleneck of Apache. and most cases with the except of an occasional slashdotting threw put is rarely past 100 connection a minute. If they really do have that sort of load where IIS speed performance is needed. They should actually rethink their strategy with having multiple servers. In the simple case if it is that popular that if one system crashed due to OS or hardware there will be a lot of trouble and profits loss. Guess what if you IT guys well need to stop thinking that Speed is the only way to improve IT. While I have seem some places with IT shops running slower equipment having better performance then ones with the fastest of everything because they were smart enough to set it up and to avoid bottlenecks. A 1 lane road going at 65 MPH for 10 miles vs. a 2 lane road going at 30 mph. Depending on the traffic the 2 lanes will allow a person to go the distance quicker then 1 lane.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Too many variables to say which one performs better, lets face it most people run IIS for ASP or ASP.NET while APACHE is used for PHP, Perl CGI, and jsp with the right modules. You really need to compare dynamic pages that look identical but are written in the relevant langs. Also performance isn't the end all on most web servers. Sure for google or amazon it is, but for an internal intranet server that is used by 500 employees I would think speed of developer deployment is more important as either server could handle the load.
Kind of like the ultimate fighting championship -- no holds barred compentition between alpha geeks to see who can webbench more.
:Windows, RHEL, FreeBSD and MacOS (Of course, the Mac group will have to work with the closest apprixmation we can manage to the x86 box going by paper specs). Their task will be to build an ecommerce site and successfully run it without getting hacked for four successive weeks, armed only the documentation provided with the operating system.
Of course this is completely irrelevant to real world usage scenarios. What we need is another data point from the other end of the spectrum. It can be like one of those reality shows. You rope four teams of ordinary folk right of the street, hand each team identical base (no OS) servers, only each team gets a different operating system
We can call it SURV1V0R.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
I work for a software company that sells a suite of server products that run on Windows/Solaris/AIX/HPUX and Red Hat Linux. We have very extensive QA where there is a team that focusses solely on performance testing. The Linux version is significantly faster than the Windows version. There is a caveat though, this applied only to Red Hat Advanced server 8. They updated the threading model or something for that one. For earlier versions of Red Hat the performance sucked for a server product.
Surprise, they enabled DisableLastAccesS on windows but did not mounted linux filesystems with noatime
noatime disables the update of the "last accesS" field of files, and improves the performance a lot for some workloads. If you check the latest article about the kernel.org servers, they found that they reduced the system load to the half by just using this option
This analisys is biased. Who cares, anyway?
orwell~> openssl speed rc4 md5 des-ede3 sha1
Doing md5 for 3s on 16 size blocks: 907943 md5's in 2.94s
Doing md5 for 3s on 64 size blocks: 807181 md5's in 2.95s
Doing md5 for 3s on 256 size blocks: 584161 md5's in 2.95s
Doing md5 for 3s on 1024 size blocks: 279567 md5's in 2.96s
Doing md5 for 3s on 8192 size blocks: 47094 md5's in 2.96s
Doing sha1 for 3s on 16 size blocks: 829742 sha1's in 2.94s
Doing sha1 for 3s on 64 size blocks: 660291 sha1's in 2.95s
Doing sha1 for 3s on 256 size blocks: 395302 sha1's in 2.92s
Doing sha1 for 3s on 1024 size blocks: 155759 sha1's in 2.96s
Doing sha1 for 3s on 8192 size blocks: 23185 sha1's in 2.96s
Doing rc4 for 3s on 16 size blocks: 13655185 rc4's in 2.92s
Doing rc4 for 3s on 64 size blocks: 4064584 rc4's in 2.93s
Doing rc4 for 3s on 256 size blocks: 1062317 rc4's in 2.93s
Doing rc4 for 3s on 1024 size blocks: 267395 rc4's in 2.93s
Doing rc4 for 3s on 8192 size blocks: 33578 rc4's in 2.94s
Doing des ede3 for 3s on 16 size blocks: 1156225 des ede3's in 2.93s
Doing des ede3 for 3s on 64 size blocks: 295752 des ede3's in 2.91s
Doing des ede3 for 3s on 256 size blocks: 75390 des ede3's in 2.95s
Doing des ede3 for 3s on 1024 size blocks: 18898 des ede3's in 2.94s
Doing des ede3 for 3s on 8192 size blocks: 2341 des ede3's in 2.92s
OpenSSL 0.9.7a Feb 19 2003
built on: Tue Oct 5 08:55:24 EDT 2004
options:bn(64,32) md2(int) rc4(idx,int) des(ptr,risc1,16,long) aes(partial) blow fish(idx)
compiler: gcc -fPIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLF CN_H -DKRB5_MIT -DOPENSSL_NO_IDEA -DOPENSSL_NO_MDC2 -DOPENSSL_NO_RC5 -DOPENSSL_N O_EC -I/usr/kerberos/include -DL_ENDIAN -DTERMIO -Wall -O2 -g -pipe -m32 -march= i686 -mtune=pentium4 -Wa,--noexecstack -DSHA1_ASM -DMD5_ASM -DRMD160_ASM
available timing options: TIMES TIMEB HZ=100 [sysconf value]
timing function used: times
The 'numbers' are in 1000s of bytes per second processed.
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
md5 4941.19k 17511.72k 50693.29k 96715.07k 130335.83k
sha1 4515.60k 14324.96k 34656.61k 53884.19k 64166.05k
rc4 74822.93k 88782.72k 92816.78k 93451.36k 93561.56k
des ede3 6313.86k 6504.51k 6542.32k 6582.16k 6567.63k
orwell~>
Before coming to the actual topic of my message, I just wanted to point out that this is a 2 year old test from April 2003. So the age of the test alone should make the results of limited value today.
However I would like to comment on how these kind of tests are performed in general.
Maybe this will outrage you, but I make the bold claim that Veritest is not to blame for the test being biased. In fact, tests like these work like follows:
- Microsoft has its own labs where they continuesly test their own products and competing products for performance and they try to find special scenarios where indeed their product is indeed better than the competition.
Once they have developped such a scenario, they pay some testing company to do the tests according to the scenario devised by Microsoft.
The testing company then just does the job its payed for and performed the tests accodring to the scenario imposed to them. Of course, they get the same results as Micrsoft and they can in all good conscience certify that in the giving scenario, the MS product is indeed faster than the competing product. So in the end, the testing company just did their job, e.g. run a Benchmark in a very particular scenario which was imposed to them. The testing company did in now way favour MS in the results. They didn't have to because the testing conditions were already biased.
Just to show how this has always been common practice, take the case of (I think it was) Netcraft a couple of years ago. They were commissioned by MS to do a benchmark which proved that Active Direcotry was faster than Novell's eDirectory. At about the same time, Novell commissioned the same company to do a benchmark that proved eDirectory to be faster than Active Directory. Of course, the testing scenarios for both cases were different.
The testing company faithfully published the 2 results. E.g. they published that their benchmark proved Active Directory to be faster than eDirectory, and the published the results proving eDirectory to be faster than Active Directory.
Marcel
Windows on a ferrari will always be faster than apache on a scooter in an unbiased timed trial.
You are in a maze of twisty little passages; all alike.
Verisoft's client list reads like a global Who's Who of the industry. AMD, Apple, Cisco, Dell, Google, Intel, IBM. Google, Nokia, Oracle, PeopleSoft, Philips, Sony, Sun. Hundreds of others. Veritest Clients. This is not a company whose reports will be casually ignored or dismissed at the enterprise level.
This way, each side may tweak their setup to the max, using all specialized knowledge, to get maximum performance. Since each side may run the optimal hardware configuration (given price restrictions), the practice of hobbling the other side by picking ill-supported hardware is prevented.
This test best conforms to the sort of thing an end user would do - pick the best bang for the buck for the budget and task at hand.
Now, this might result in a dual Itanium server (Windows) being benchmarked against a dual Power server (Linux) (or some other comparison), but that is "fair" in that both sides are running on the same COST hardware.
True, each side might "release" a new (service pack|set of RPMs) for the purposes of the test, but as long as those releases are publicly available, who cares? We all benefit from the improvement of the code.
www.eFax.com are spammers
Ask yourself why you chose the webserver you chose and the particular platform it runs on. Was it because of speed alone? Most likely not. And if you do need more speed than standard Apache on Linux gives you, what would you do? Buy more servers? Perhaps. It probably costs less to have 3x the number of Linux servers measured in licence cost, in time spent developing for them and keeping them running.
Of course, like everything else: it depends.
Besides, if you need more performance: design better web apps. Seriously. Most web apps have bottlenecks that are due to bad design and do not reflect the performance potential of the underlying technology at all. I am constantly amazed at the sub-optimal solutions developers pick when solving a particular problem. Scaling up systems isn't really all that magic. It is mainly about knowing what problem you are trying to solve.
I'm not really interested in these benchmarks anymore. I was perhaps 10 years ago, when webservers weren't that good and there were some real performance gains available through relatively simple tricks. But now most any webserver will do because web sites are usually constrained by other factors. Like bandwith and the running times of their web apps.
Microsoft lost this round of web server platform wars. Perhaps Longhorn will put a dent in the market, but you'll never know. Microsoft have this fantastic way of creating a lot of buzz for their OS releases, and when they finally materialize it isn't really that much of a quantum leap.
While using identical hardware seems like a reasonable level playing field, I think it would be better to give both teams an equal budget for hardware. Most companies trying to make a decison are going to look at their budget and see how much web server they can get as a system. Not I'm going to buy X hardware and Y software, which might not play particularly well together.
If you want a more realistic test, and to tilt the field in favor of open source, make the budget for hardware and software.
You also need to factor in the test clients. IE still has the biggest market share, so the test clients should behave like IE a proportional amount of time. But it should also behave like FF/Moz/Op/lynx a proportional amount of time.
The unedited text of the study:
Windows 2003 web server performs up to 300% higher DDoS throughput than Red Hat Linux running with Apache
Knowing my 2 cents amounts to nothing, I think an international committee could be formed to oversee a challenge match.
Isn't it time for server Olympics?
Lost in space at an early age. Survived the vacuum. Now rebuilding castle in air.
The cut and paste in a plain ol'timey X window, though is far superior for my heavy use. It is almost as effecient (operator wise) as 'drag and drop', just one more mouse click. Now the MS rip variation of cut and paste takes much more interaction per cut and paste operation multiplying out to much more and slower work for the user.
I have to applaud the way you take a positive stance and look at how apache can be improved. I expect efforts in that direction form an ongoing part of apache development, but the positive attitude is appreciated. It's just a bit sad that your post reads as an endorsemnt of a blatant piece of paid-for propaganda
Don't let THEM immanentize the Eschaton!
Some people still do - the ones that aren't working with open source, and want to sell a product without giving away their source code. There are also people who want to obfuscate their web code to protect their job. Hard to be indispensible when they can fire you and bring in a monkey to make minor layout changes after you did all the initial hard yards.
Yay me!
They shut off access logging in IIS. As far as I could see, they left logging on for Red Hat. This means that lots of disk writes were being generated on Linux but not on Windows. As http request volume goes up in their tests, the RAID write-cache could eventually fill up (only under Linux), at which point the webserver starts blocking while waiting for disk I/O to complete.
Figures that right after submitting this I see that they turned off access logging in Apache. Doh!
---------------------------------------------
SERENITY NOW!!!!!!!!!!!!!!!!
the problem with web server benchmarks is that tuning of the web server will change them immensely. Apache, for instance, can be tuned to cache all data served out of memory, making it extremely fast. Furthermore, you can use kernel mods under linux to further accelerate static page serving. So once it amounts to reading data from memory and writing it to a socket, you're looking at TCP stack tunings and scheduler tunings in the OS.
Unless one OS has a fatal flaw in its TCP Stack or Scheduler, they should be able to approximate eachothers performance with similar tuning parameters. So what you are really testing is TCP performance, no?
The budget has to buy software, hardware and setup labor.
This eliminates the problem of "that hardware favors Microsoft" or "that team had better engineers". It all comes down to money and value.
Of course the competition would need to state up front exactly how performance would be measured and how the various different tasks (static pages, cgi, etc.) would be weighted to come up with any overall scores. That would dictate the design choices made by each team.
One simple rule for its versus it's
I think it is somewhat justifiable to discount this study on it face. How many times do you stick your finger in the flame, before you learn that it is not a good idea. Same thing with MSFT sponsored studies,
They have been found wanting so many times, that most people don't bother with any investigation.
I'm more a fan of the keyboard than the mouse, so the MS way is just as good for me, and Windows applications are typically a lot better overall in terms of keyboard accessibility than X applications.
However, I was referring more to the meaningful translation of complex data when copying/pasting, dynamic updating of linked data, etc. I can do the same thing with command-line UNIX software by using a makefile to regenerate the final output after changing source data, but X applications, in my experience, can't compare to Windows applications here (and on Windows, MS applications are typically better in this respect than non-MS ones).
Apache can saturate your 100MB ethernet (what you will get in most colos) quite easily. All other traffic is queued. So most of this talk is pointless drivel. The world's most visited sites use Apache, so clearly it works.
I wonder if they're running Linux. You'd think by now they'd have read their own study and found that Windows 2003 Server is faster. ;)
Give it a rest.
We're switching to Microsoft WIndow 2003 tomorrow!!!
Hey - totally off topic, but that IS a great book. If you find an online version that is unedited (more importantly, unannotated) - please share. Reading the annotated version is tantamount to trying to eat a nice dinner and having the phone ring every 10 minutes.
As for the Linux vs Windows performance debate - sounds like a paid shill that intentionally tweaked the comparison so Linux would lose - but at least they are being honest about it.
Glonoinha the MebiByte Slayer
Microsoft is trying to test apache 2 running on linux. Large isps and yahoo use FreeBSD. Why not prove its faster than apache2 on freebsd. That would impress me.
.12)
.NET vs Mono on linux and windows (doesn't work right in BSD) .NET using all the UNIX stuff for java.
As others have pointed out, Redhat is not the best Linux platform for a test of this nature. How about using a custom compiled (or even stock) 2.6.x kernel with and without an apache module. ( i think you guys have one in the linux camp) IIS uses a kernel module so that is fair. Hell why not use Gentoo, Debian or Suse with a recent kernel?
I personally think FreeBSD is the best for a webserver, but lets at least see a fair Linux benchmark. Why not do a real benchmark..
Linux 2.6.10 (or
Linux 2.4.x
FreeBSD 5.3 or 5.4 beta
NetBSD 2.0
Darwin or Mac OS X Server
Solaris
SCO UnixWare (just to laugh)
Windows 2003 server
Windows 2000 server
Do several tests...
static content
php on all platforms
asp on all platforms (chilisoft for linux/solaris and in emulation mode on bsd)
Maybe java vs
Now that would be an interesting benchmark. I could make a real purchasing decision then.
Even if microsoft looses, they still have the fastest enterprise web app product for development time. PHP, Mono and java might compete some day.
Obviously, single and multi cpu (or cores) should be used with and without hyperthreading. Also, to be fair the SCSI or SATA controller should be equally stable in the oses and preferably a network card that works ok in all (Intel chipset?)
It is not necessarily anticonstitutional. I.e. an employer can fire you for what you say to customers. So commercial speech is not protected under the 1st ammendment. (IANAL, of course).
What is disturbing, however, about these no benchmark clauses is that they are likely to be anti-consumer rights and anti-competitive. I.e. they are specifically designed to enforce a "buyer beware" marketplace where one can simply not get good data on reasonable performance of competing products. The ability of a company to regulate what another company says about their products as a condition just to use them seems to me to be dangerous to the idea of a free market.
LedgerSMB: Open source Accounting/ERP
Apache on Linux can run on a 486 w/ 16MB RAM. Windows 2k3 can't. Why? Because it is a resource hog.
Its not the only reason this report is BS but it is highly unlikely that a setup that can run on a really old system would run slower on modern hardware than windows 2003 on the same hardware.
The "big players" can contribute cash, but not hardware. It is too easy for them to contribute hardware specifically enhanced for their product.
#1. Each team gets X dollars and no restrictions on what it can buy. After all, that should be how businesses run their shops. We aren't comparing hardware, but total systems.
#2. Each team must purchase the software off the shelf.
#3. No team is allowed to recompile anything or to use any drivers, etc not available from a public server for the past 12 months. This might sound like a bad deal for Linux, but it will also stop Microsoft from re-writing the drivers. Again, most companies do not have access to that level of expertise so that won't be allowed.
#4. Each tweak or configuration setting must be documented and a reference for it shown on a public website or manual. Again, businesses only know what they can read.
#5. At the end of the competition, the other teams will critique each team's configuration. We've all seen the "tests" where Windows is running on a RAID 0 array which is beyond stupid for real production work.
That way, each team can deploy the best system they can think of for the test. I'm sure you all remember MindCraft and their massive single server "test" for webservers when anyone else would have run multiple cheaper servers and gotten higher throughput.
So, a test in run and the Windows team buys the biggest single system they can afford for the money. While the Linux team fields a dozen boxes booting from CD and one storage box.
Which system would be "better"?
Which system would be faster? Would that be the same answer under different loads?
Which system would be easier to maintain?
Which system would have higher uptime?
Which system would be easier to scale up?
I'd like to see them benchmark the 2003 IIS server against a stock OS X server. on equal machines, or as close to equal as possible. While sales people use stuff like this. A slightly slower but stable webserver will beat out whatever speed machine one can come up with. And we all know you can build multiple redundant unix servers for the cost of MS licensing.
And just an opinion -- I don't think Linux would have got so popular if it wasn't for its anti-microsoft/windows (replacement) approach.
If that were the case, Apple would be very very happy right now. There are many affirmative reasons for liking Linux (openness, quality, average ethics). The only reason people make this mistake is that a lot of those are also affirmative reasons to loathe Microsoft.
For the love of God, please learn to spell "ridiculous"!!!
Tune all you want, apache will not match the performance of fast web servers like zeus and lighttpd. And it can't handle as many concurrent connections either. This benchmark is not a big deal, and its not suprising. Apache is not meant to be high performing, it is increidbly flexible and configurable, and performs well enough for most tasks. If you need higher performance use a webserver designed just for high performance.
"- -George W. Bush The Great Divider"
George W. Bush, The Despicable Divider
Apache isn't meant to be the fastest webserver out there. Its a very flexible, configurable server with good enough performance. IIS is designed for high performance, and even speeds things up by using a kernel driver to do part of the request. If you want a more reasonable comparison, try comparing IIS to something designed to be fast like lighttpd.
I'm guessing they don't factor in all the downtime while you patch the Windows OS for the latest worm or virus, or reinstall the OS when it gets infected.
Then I compare them with offerings like Mac OS X, the BSD's and Linux and wonder, how on Earth someone can say, "I like Microsoft".
They have been abused so long by their captors that they have actually grown to like them.
http://www.hackiis6.com/default.htm
Shouldn't prove to be much of a problem then. Oh wait, nobody would ever waste their secret knowledge of vulnerabilities on a mere x-box... by.. bi.. bass... bias! there we go, I think that might have something to do with it.
"Windows 2003 web server performs up to 300% higher throughput than Red Hat Linux running with Apache."
They were testing for a real-world enviroment. Apache would be serving large pages with plenty of images. Windows Server would be serving a single line of text to it's thousands of visitors.
"Hacked by Chinese!" =P
Windows has detected an undetectable error.
stat jokes. Isn't it computer programmers use logic, physicists understand logic and statisticians cheat logic?
What it comes down to in the end, is that Apache on Linux performs to my liking. Let's not nit pick about which one is faster. From a security and administratability stand point, Apache on Linux blows IIS on Windows out of the water.
This seems like Microsoft is advertising to the inexperienced or fly by night sys admins.
Microsoft can't get a reasonable research done without paying for it, and if they pay for it the people they pay will pretty much say anything they want to. Microsoft has something to lose if the test tells anybody they are not the all-over winners, so the test will show that they are and thus be totally useless. Most, if not all, Microsoft-funded researchers I've seen reports from always report between 25 and 500% profit from using Microsoft software, with the top number being implied as the generic profit.
.NET). They then report the average of the better scores as probable profit, which isn't entirely unbiased but to be expected.
On the other hand, Apache (and the other OSS guys) also have something to lose by showing a solely bad report. They don't have anything to fear from partly bad reports or good reports. They can be largely trusted, but keep in mind that there is a good chance that they're completely sucking in one or more departments. I've seen these papers include one or two benchmarks that show >100% profit but they can usually be claimed to a single obvious cause (say, the IIS server running an apache compatibility layer or Apache running some win32 dll to get something working, such as
Main conclusion I can draw from the system of capitalization is that you can not trust anybody. If somebody has ANYTHING to lose at a certain result, the result will never show and the test will not be reliable. The only people that are reliable are those that don't have anything to lose but their life. Yet, these people only work for public broadcasting corporations under a do-what-you-like agreement so they could be honest. And still you can't trust them entirely, since there is something to gain from a positive result...
How to figure out which is better? Try them both (fsck illegal software, just friggin download it and try which is better, you are the only one that can be truly honest to you, not counting schizofrenics) and roll out the better performing one after paying for the licensing costs.
Oh, and donate to the poor-Microsoft-fund because the GPL is a price-fixing scheme that's designed to make Microsoft run out of business (explanation: Microsoft does what everybody did 10 years ago. If RMS can give it away 3 years after they did it Microsoft has nothing to do anymore).
1. They only used HTTP 1.0 requests, which isn't normal for high-end performance testing
2. They completed testing on April 30th, 2003, a year before the report was published, and a year has since passed
3. They used the 2.4 kernel instead of the 2.6 now available
4. They used an early Apache 2.0 configuration which I believe was known to be somewhat slower than the apache 1.3 code at that time.
Draw your own conclusions
quote from report..
...so, basically, they couldn't figure out how to configure the system properly, sent in a support call, and didn't get a reply, so published anyway... maybe trying to make the point that support might not be so good with RH... be good to know what happened to that support call..
;)
"
When testing with eight processors, the static test results
actually decreased approximately 23 percent compared to the results generated using TUX with Red Hat
Linux Advanced Server 2.1 using four processors.
We double checked the TUX and operating system configuration and ran additional tests to try and resolve
this issue, but were ultimately unsuccessful. We submitted a formal support request with Red Hat Technical
Support on April 8th, 2003 regarding this issue with TUX and provided problem and configuration details. On
April 9th, 2003 we received an indication that out request had been escalated to the senior technical support
staff. On April 30th we had still not received a response from Red Hat Technical Support that offered any
options for resolving this issue. By this time, we were required to return the servers used during the testing to
Hewlett-Packard. This meant that even if Red Hat Technical Support had responded with options for resolving
this issue, we would no longer have the hardware necessary to conduct additional testing. Therefore, we
published the existing numbers shown in this report.
"
but then, just maybe, they could have gotten a linux guy to have configured their systems for them
I don't care if Windows is faster, I won't use it. Just because its faster is some benchmark doesn't mean jack because I would still have to administrate the POS. I'd rather just buy a faster system.
It is unfair to go until the first moment of downtime. Who knows, one server might last a month longer than the other before the first bit of downtime, but could still have more total downtime. It would be fairer to count the total amount of downtime for the one-year period. Especially if an important patch is released for one OS which requires a reboot.
First, look at the freaking date before submitting something as "News". The PDF is dated May 5, 2004 (not 2005).
Here are some observations:
- The Windows version tested was 2k3 pre-release. To be fair, VeriTest should have tested RHEL 3 and RHL 9. Otherwise, it should have tested against Windows 2k server.
- The IIS version tested was 6.0. To be fair, VeriTest should have tested with the current Apache releases at the time (1.3.29 and 2.0.48/9) and the current Tux release included in RHEL 3.
Other notes:
- The Linux kernel rev for RHEL 2.1 is 2.4.9-e3. IBM benchmarked web serving performance for 2.6 vs. 2.4 and the results are huge performance gains by using a 2.6 kernel.
- The filesystem used was ext3. ReiserFS is a faster filesystem for filesystem, although it uses more CPU.
- There was minimal performance tuning to Linux or Apache. TUX was performance tuned. The testers attempted to reverse-engineer settings for Apache.
Bottom line: This survey isn't much better than the Mindcraft survey done several years ago. It didn't tune Linux at all, and received tuning help and funding from Microsoft.
Is one that much faster than the other that it causes the other server to become unuseable? Both work, and faster than their predecessors. Although you'll spend the majority of your time worrying over MS2003 security, both will work. Benchmark studies like this are just contests to see who has the bigger hard-on... is it better to have a 12-incher, or a 12 1/4-incher?
I haven't had to upgrade my email/web server hardware in 10 years, thanks to Linux/Apache.
I can't imagine W2K3 even installing on such hardware.
No where in the "review" does it mention what kernel version is used by RHEL 2.1 or RH8. I know RH9 came with a stock 2.4.9 or something like that so I'm assuming RH8 is probably even older like a 2.2 kernel and RHEL 2.1 is a 2.4 kernel. This is like comparing win98 performance to winxp performance. I would say that the only fair benchmark between IIS and apache (and not linux) would be to run the same tests on IIS and apache both on win2k3 server (yes apache is available for win). If you want to benchmark operating system performance then go bench the same programs on win2k3 and the same system running the latest kernel.
Decent hardware (Mice)
Excel (The spreadsheet, that is)
Providing competition for the free operative systems.
Providing inspiration for http://ubersoft.net/, among others.
fnord(gazonk, foo, bar, baz, bletch, thud, grunt)
Nice to see Tux is still developed, but these days does it do much more than
khttpd now in the kernel?
The content acceleration on 2K3 seems to work with dynamic content too and they
SEEM to have some pretty decent caching (a bit like how we're stuck with Turck
or Zend). All of this is stuff you need to go out and get on Linux, but just
check a checkbox on 2K3 if it's not enabled by default. This is probably
inflating the benchmarks a LOT on the 2K3 side too.
Neko
The first thing you should do as a web administrator on your P4 class RH box running Apache is tune the following settings. Whether you use a tool, or edit directly depends on what kind of support you expect when you mess it up. The below is one of many possible solutions depending on available memory, processor speed, SMP capabilities/kernel version, and whether you anticipate a high rate of connections or a larger amount of CGI type work:
- StartServers - typical default is 2. If you're half serious about your web server being able to respond to requests from the completion of startup, try 10. That's just to get the ball rolling.
- MaxClients - 150 is a typical value. Try 1000 so you're prepared for incoming connections.
- MinSpareThreads - 25 may be ok usually, but if you get a surge of benchmark style incoming connections, you're bound to be caught out not ready to accept some. Try 100, or 200 if you have a little memory to burn.
- MaxSpareThreads - 75 might seem decent, your SMP box will be much more responsive with 200, perhaps a better idea being 500. Even 2000 if you can see memory usage isn't a problem.
- ThreadsPerChild - 25 is ok in my book. You probably don't want each process responsible for too much, especially if they do uninterruptible things. But you may be able to up it for a purely static site.
Remember, if you are running a web server, you should be tuning it to use all available resources, without thrashing your swap space. Look at the numbers and tune it. Expect any modern PC running Linux/Apache to take 5000 static hits per second sustained on an SMP box without breaking a sweat. If during a peak time you can see a resource underutilised, then tweak and tune it - any good admin should be able to squeeze orders of magnitude better performance out of it over the default config. Using half the memory and bugger all swap? Up the threads by 50% or more...Once you have your web serving building block optimised, virtual services can then be used to load-balance traffic for extreme peaks, and taking boxes down for essential maintenance without impacting on the users. That's the cool thing about
(end of previous :)
IANAL, but... aAn NDA would prevent you from disclosing the inner working of a piece of hardware/software/etc... but I don't think it limits you as far as the product's performance is concerned.
I might sign an NDA saying that I won't give details about "Windows 2009 Ultra Secret Edition" and what features it has, but I could probably still make generic comments like "the current version crashes a lot or is slow on common tasks" just the same as I could say "wow, it' a lot faster and the new graphics options are incredible" (without actually saying what the options are).
What kind of testing is "there testing", I've never even heard of here testing.
Oh maybe you wanted to say their testing. Go back to sixth grade and pay attention this time.
[eak@www ~]$ openssl speed rc4 md5 des-ede3 sha1
Doing md5 for 3s on 16 size blocks: 501077 md5's in 3.00s
Doing md5 for 3s on 64 size blocks: 447089 md5's in 3.00s
Doing md5 for 3s on 256 size blocks: 328345 md5's in 3.00s
Doing md5 for 3s on 1024 size blocks: 161911 md5's in 3.00s
Doing md5 for 3s on 8192 size blocks: 28091 md5's in 3.00s
Doing sha1 for 3s on 16 size blocks: 450041 sha1's in 3.00s
Doing sha1 for 3s on 64 size blocks: 360994 sha1's in 3.00s
Doing sha1 for 3s on 256 size blocks: 227294 sha1's in 3.00s
Doing sha1 for 3s on 1024 size blocks: 92109 sha1's in 3.00s
Doing sha1 for 3s on 8192 size blocks: 14104 sha1's in 3.00s
Doing rc4 for 3s on 16 size blocks: 8473853 rc4's in 3.00s
Doing rc4 for 3s on 64 size blocks: 2499745 rc4's in 3.01s
Doing rc4 for 3s on 256 size blocks: 650560 rc4's in 3.00s
Doing rc4 for 3s on 1024 size blocks: 163453 rc4's in 3.00s
Doing rc4 for 3s on 8192 size blocks: 20386 rc4's in 3.00s
Doing des ede3 for 3s on 16 size blocks: 711139 des ede3's in 3.00s
Doing des ede3 for 3s on 64 size blocks: 183706 des ede3's in 3.00s
Doing des ede3 for 3s on 256 size blocks: 46336 des ede3's in 3.00s
Doing des ede3 for 3s on 1024 size blocks: 11549 des ede3's in 3.00s
Doing des ede3 for 3s on 8192 size blocks: 1445 des ede3's in 3.00s
OpenSSL 0.9.7a Feb 19 2003
built on: Tue Oct 5 08:55:24 EDT 2004
options:bn(64,32) md2(int) rc4(idx,int) des(ptr,risc1,16,long) aes(partial) blowfish(idx)
compiler: gcc -fPIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DKRB5_MIT -DOPENSSL_NO_IDEA -DOPENSSL_NO_MDC2 -DOPENSSL_NO_RC5 -DOPENSSL_NO_EC -I/usr/kerberos/include -DL_ENDIAN -DTERMIO -Wall -O2 -g -pipe -m32 -march=i686 -mtune=pentium4 -Wa,--noexecstack -DSHA1_ASM -DMD5_ASM -DRMD160_ASM
available timing options: TIMES TIMEB HZ=100 [sysconf value]
timing function used: times
The 'numbers' are in 1000s of bytes per second processed.
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
md5 2672.41k 9537.90k 28018.77k 55265.62k 76707.16k
sha1 2400.22k 7701.21k 19395.75k 31439.87k 38513.32k
rc4 45193.88k 53150.72k 55514.45k 55791.96k 55667.37k
des ede3 3792.74k 3919.06k 3954.01k 3942.06k 3945.81k
Armaments, 2-9-21 And Saint Attila raised the hand grenade up on high, saying, 'O Lord, bless this Thy hand grenade' N
I want to see one of these three vendors take their best team of Windows IIS Support and professional services engineers and their best team of Linux support and PS engineers and do a real head to head challenge.
Give each team your best hardware. 2, 4 and 8 CPU systems. Fiberchannel storage arrays configured for maximum web server performance in a customer production configuration (Raid5, BCVs, massive spindle luns, 8K, 16K and 32K block size volumes.
Leverage your partnership with LoadRunner. Build a massive load generation farm. Full GbE on generators and servers. Jumbo frames might be interesting. Fully document the tweaks from hell to eek out the maximum web and application server performance from each system.
Use the latest and greatest version, fully patched of each OS. Leverage your Microsoft and Linux distribution partnerships. Hell lets make this really interesting and bring in X86 Solaris and Apple, too! Get Microsoft to send a team of 200 and the Linux distribution vendors to send 1 guy, a trunk of O'Reilley manuals and his buddy from the Apache team.
1 week to prep the web server hardware. All tweaks, tunings and configuration changes must be documented and submitted to the other team on day 6. Either team can take advantage of tricks discovered by the other for tweaking hardware. Prepare your server for a 20 million hit load test.
The exact same HTML, XML, PHP, java and javascript will be loaded on each server in as similar as possible directory structure. No vendor specific coding tricks or proprietary BS. If I can't properly view the page in 3 of 4 web browsers (Explorer, NetScape, Mozilla and Opera - or more if you feel its necessary) it gets removed.
Then bring in Load Runner. Let them create the hairiest pounding load generator that hits every page, every CGI, every java applet concurrently from hundreds and thousands of clients.
The first to do so will be the god of Linux. All those slunking down to hide behind their Microsoft coat tails just lost the race.
I dare ya! I double dog dare ya!
Armaments, 2-9-21 And Saint Attila raised the hand grenade up on high, saying, 'O Lord, bless this Thy hand grenade' N
If they did the tests with RHEL3, it might be right.
We've been using a couple SunFire's (dual xeon 3.2ghz, 5gb memory, scsi disks) with gentoo and rhel3, and rhel was almost 3 times slower.
Upgrading to RHEL4 changed things though, it's almost on par with Gentoo, but a lil wee bit slower.
That's what you get with "enterprise certified bullshit".
They put together the fastest computer possible with the most memory possible, the fastest hard drive possible, and all the fastest stuff possible. Then, they installed Windows 2003, and they had their programmers who know this thing inside and out remove all the parts of the OS that were not needed, along with configuring it for the maximum possible performance. Then they tested it and recorded the benchmark.
Then, they got an 8086 computer and compiled Bochs to run on it. In Bochs, they set up an environment that looks like the identical computer they used to run Windows 2003. Because this 8086 has only a few kilobytes of memory, they actually stored the information on tape, and used the serial port to access that data. Then, they ran the test. Sure enough, Red Hat / Apache running on the Bochs running on the 8086 with no memory and all information being accessed through tape through the serial port, the Windows 2003 installation ran 300% the speed of the Red Hat. This isn't quite such bad news, though, because it shows not how slow Red Hat is running in such a screwed up scenerio, but rather how slow Windows is running on ideal hardware.
I mean come on. From the article the compared enviroments are:
Windows Server 2003 (their latest)
IIS 6.0 (the latest)
vs.
Apache 1.3.23 What is that, like 2 - 3 years old now?
RedHat 2.1 . Last I checked, they were on 3.0. 2.1 is also 2-3 years old.
Now, if they had RedHat3.0 with Apache 2, well, maybe I'd read further. But comparing a 2-3 year old configuration with the latest and greatest from your company seems to make the whole premise invalid.
You know, to deliver apps effeciently you really need to tune the kernel.
This study runs a DL760 from HP. The DL760 ships with REdhat AS 4 or Windows Server 2003. Redhat 2.1 is not even an option.