Slashdot Mirror
Linux Beats Win2000 In SpecWeb 2000
PraveenS writes: "While not conclusive, the SPEC group released benchmarks for a variety of systems submitted by various manufacturers (i.e. Dell, Compaq, HP, etc...) and tested their Web-serving capability. Two very similar machines from Dell, one loaded with Linux and the other with Win2000 had very different results; Linux beat Win2000 by a factor of almost 3 . Here's a synopsis of the results from LinuxToday. The actual spec benchmarks are available here for Win2000 and here for Linux."
As Marty of LinuxToday puts it, though, "What does this mean? In the real world, probably not as much as it would seem. Benchmarks in general are typically set up in an ideal environment. Real world environments tend to be quite different. However, this does indicate that Linux is moving in the right direction."
Zoran points out that "[o]ther current SPECweb99 results can be found here." They make an interesting comparison.
Isn't this the khttpd that was supposed to make it into kernel 2.4 versions (which I thought was a fairly crazy idea; a kernel serving http) ?
Religion is what happens when nature strikes and groupthink goes wrong.
Gigabit..... sigh, what can I say about it other then I personally have had nothing but problems with it in both NT and Linux. I get a whopping 3600K/sec when I'm on 100mbit and when we switched a few machines to 1000mbit fiber we get 3800K/sec. Great we just spent ungodly amounts of money on a nice Cisco 5300(?) Catalyst router and 450.00 a piece on some network cards. I think the problem is in how the router is set up though. I'm assured by a few people that it really isn't supposed to be that slow ;-)
This is Oracle's '$1000000 prize' trick, any hardware you like against any hardware we like. Of course, MS managed to do TPC/C on a shared-nothing cluster, making a mockery of the whole thing, and pushing TPC/C even further in the direction of a benchmark of nothing more than the amount of money a company is willing to spend.
Speaking of which, I've used an S80 (the M80's big brother), there is NO WAY that any 4 CPU PC is 50% faster than an 8 CPU RS/6000 M80 on any comperable benchmark. This raises a large question-mark over the whole result.
Redhat/Dell have done nobody any favours by publishing such a clearly unrepresentitive benchmark result. It's Mindcraft all over again, this time from the other side, redoubled.
If you had bothered to read more of the site, you would have noticed that those results are all submitted by vendors with interests in getting good numbers. If Sun wants to enter a Sparc/Solaris combo, they can do that. If Apple for some reason decided it was in the HTTP server business (which it isn't (yet?) by any stretch of the imagination), it can run the test suite and submit the results.
Gates' Law: Every 18 months, the speed of software halves.
'TUX' comes from 'Threaded linUX webserver', and is a kernel-space HTTP subsystem. TUX was written by Red Hat and is based on the 2.4 kernel series. TUX is under the GPL and will be released in a couple of weeks. TUX's main goal is to enable high-performance webserving on Linux, and while it's not as feature-full as Apache, TUX is a 'full fledged' HTTP/1.1 webserver supporting HTTP/1.1 persistent (keepalive) connections, pipelining, CGI execution, logging, virtual hosting, various forms of modules, and many other webserver features. TUX modules can be user-space or kernel-space.
The SPECweb99 test was done with a user-space module, the source code can be found
here. We expect TUX to be integrated into Apache 2.0 or 3.0, as TUX's user-space kernel-space API is capable of supporting a mixed Apache/TUX webspace.
TUX uses a 'object cache' which is much more than a simple 'static cache'. TUX objects can be freely embedded in other web replies, and can be used by modules, including CGIs. You can 'mix' dynamically generated and static content freely.
While written by Red Hat, TUX relies on many scalability advances in the 2.4 kernel done also by kernel hackers from SuSE, Mandrake and the Linux Community as a whole. TUX is not one single piece of technology, rather a final product that 'connects the dots' and proves the scalability of Linux's high end features. I'd especially like to highlight the role of extreme TCP/IP networking scalability in 2.4, which was a many months effort lead by David Miller and Alexey Kuznetsov. We'd also like to acknowledge the pioneering role of khttpd - while TUX is independent of khttpd, it was an important experiment we learned alot from.
Other 2.4 kernel advances TUX uses are: async networking and disk IO, wake-one scheduling, interrupt binding, process affinity (not yet merged patch), per-CPU allocation pools (not yet merged patch), big file support (the TUX logfile can get bigger than 5GB during SPECweb99 runs), highmem support, various VFS enhancements (thanks Al Viro), the new IO-scheduler done by SuSE folks, buffer/pagecache scalability and many many other Linux features.
But businesses can.
True, but why pay the software licenses when you don't have to? A friend at work got a new laptop a few months ago. The cost of the software licenses was about half the cost ($1000-2000) of the complete system. Multiply that by hundreds or thousands of employees every two or three years and you start forking over a considerable amount of cash. IMHO, the amount would have been better spent funding a free software project for the same task.the good ground has been paved over by suicidal maniacs
Win2000: Log Mode W3C: Extended Log File Format
Linux: Binary CLF
Looks to me that binary logging can be much faster then logging in the W3C extended logfile format. Please correct me if I'm mistaken, but I think this can really cause rather different results
Every expression is true, for a given value of 'true'
I guess it really depends. First of all I probably would not use apache for a benchmark aolserver is much faster. Also aolserver has persistent database connections (so does php).
There are several issues.
1) Database speed. In a typical web based environment or read mostly-write rarely mysql would stomp on sql server introduce frequent writes and the reverse would occur. For a transactional environment try interbase it's fast and robust and stable as hell.
2) Web server speed. This is a close one but I think aolserver might edge IIS either way it will be a close call.
3) Middleware. This is where it gets very very tricky. If you are writing simple aps pages using ADO to open and close databases AOL server will trounce asp so will php. In order to write scalable ASP pages you will need to utilize MTS heavily. You will need to write COM objects either in C++ or VB and register them with MTS. In other words you will need to double or triple your developement time and run into insane debugging problems. This is where the Aolserver or php environment really shines. Automatic database pooling and very parid developement easily pays for whatever performance hit you may take.
Of course the real solution may be to use JAVA servlets. For complex web sites J2EE is a compelling solution and much easier to program then DCOM.
War is necrophilia.
Well, nothing major will change in an instant. But improvements are creeping up on us in a steady pace, day by day. [I am level 26 and have a 4 point blizzard (3 point added by magic staff). Mephisto were a piece of cake]
MS says they've "improved" the parsing speed of ASP vs HTML in W2K until there is no/little difference between them. Look up on MSDN / aspwire for more info. Not that your advice is not sound or correct. I'd still do the same (if/when) I migrate.
Dunno if the've stopped W3SVC crashing every two minutes tho!
All spelling mistakes are in my mind and are faithfully reproduced by my fingers
Benchmarks have a tendancy to prove what they are supposed to prove. The results that are gotten from benchmarks, are the results that they wanted to get.
Go figure there were no results for real Unix flavors.
But not very realistic. I mean, quite well removed from a good test of real life performance.
perl -e 'fork||print for split//,"hahahaha"'
> Linux zealots scream bloody murder and inspect the process with a microscope. Someone else does a benchmark that shows Linux 3 times faster than Win 2k, and they are content that the Mindcraft fiasco has been avenged.
Well, it could be that we notice that a standard benchmark was used rather than one tailored by a company with an axe to grind. Or it could be that the benchmarks were submitted by hardware vendors, whose primary interest is in making their hardware look good (i.e., it's really hard to imagine Dell fudging a benchmark to make Linux look better than Windows). Or it could be that c't already told us how Linux and NT measured up on more equitable benchmarks. Or it could be that Microsoft's own tests showed W2K performing worse than NT on systems with > 4Mb of RAM. Or it could be that testers have been saying that W2K needs +300 MHz in hardware to perform as "well" as NT did.
In short, there's no reason for surprise at all. This benchmark is only quantifying what the attentive already knew qualitatively. If there are flaws with the benchmark, they almost certainly won't be enough to tilt it against what we already knew; if they do, we'll air our suspicions again.
I do agree that it's still a benchmark, and is therefore susceptible to all the follies associated with benchmarks. But at least this one wasn't obviously rigged.
--
Sheesh, evil *and* a jerk. -- Jade
For Linux they set the backlog at 3000. For W2k it's at 1000. Anyone see the difference? AFAIK, W2k can have a higher backlog, or even a dynamic backlog. I'd like to see a test where the backlogs are the same. Then there would actually be similar simultaneous connection counts! Right now, those numbers mean little if they are being compared.
For instance, on the Windows side you might have an 8 way xeon with 2 gigs of RAM. On the Linux side you might have (for instance) an S390 with a terabyte or two of RAM. Then just start loading them down with network clients until they start to stagger.
I'd be interested in the oucome...
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
Hasn't Linux always been moving in the right direction?
___
A requirement of creativity is that it contributes
to change. Creativity keeps the creator alive.
___
I'm an exhibit on the mounted animal nature trail.
"also, how do you figure that Macs are any less "desinged for this" than x86 boxes"
Macs are first and foremost designed to present the most intuitive and consistent interface to the user sitting at the desktop. x86 boxes are a hodgepodge of different components that people pick and choose from depending on the purpose of the computer.
That said, the dual processor Dell boxes benched here are a world away from G-4's and Athlon desktops. We use a Dell file server here in the office, and they are simply an amazing piece of machinery. Everything -- the motherboard, the case, is engineered for optimum server performance. (There is a even a drag-racer style fan duct just to cool the processors!). The G-4 is a beautiful piece of engineering, but it's made for digital video, not as server.
Comment removed based on user account deletion
This boils down to: TUX is in the kernel plus some modules. Great for speed. Bad for security and stability.
--
Linux MAPI Server!
http://www.openone.com/software/MailOne/
(Exchange Migration HOWTO coming soon)
> In any case, if it's in kernelspace, it's most likely not a full-featured HTTP server like Apache, Zeus, IIS. So it can spit out static pages as fast as you'd possibly need. Big deal.
Yeah, the more you look at it, the more it looks like they were just doing a Mindcraft '00. The most memorable joke then was that the only sites to use big static pages on such expensive hardware and matching network bandwidth would be the more profitable p0rn sites.
If the configuration was bullshit then, it's still bullshit now. Someone please tell me that Red Hat hasn't spent the past year tweaking things just to beat Microsoft on Mindcraft '00. I guess these things play well with the PHB crowd, but surely there are better things Red Hat could be doing with their time and money.
--
Sheesh, evil *and* a jerk. -- Jade
yes, but if I recall correctly, the chose to use a RAID controller that was *barely* supported for Linux (beta quality?). There was also an issue with the network cards but I forgot what. And a huge amount of RAM (too much for Linux at the time).
The point is: the hardware was specially selected to favor Windows.
That's just to address the hardware. The software tweaks and configurations they did was just a joke - but the second round of tests addressed some of the software issues.
You are confusing two completely different architectural concepts.
"threads" (which get created) and "processes" (which get forked) are 'context of execution' entities. Linux has both, TUX 1.0 uses both.
A "threaded TCP/IP stack" is a slightly mis-named thing, it means "SMP-threaded TCP/IP-stack", which in turn means that the TCP/IP stack has been "SMP-deserialized" (in Windows speak) - TCP/IP code on different CPUs can execute in parallel without any interlock/big-kernel-lock overhead or other serialization.
A 'threaded TCP/IP stack' has no connection whatsoever to a 'threads'.
FYI, the Linux TCP/IP stack was completely redesigned and deserialized during the 2.3 kernel cycle, this redesign/deserialization was done by David Miller and Alexey Kuznetsov. The TUX webserver of course relies on the deserialization heavily, but this is not the only architectural element TUX relies on.
There's nothing inherent in the design of the G3/G4 that limits it to use by graphic artists. Anyone that knows anything about chip architecture knows that it's basic design is leaner, cleaner, and meaner than x86 because it doesn't suffer from having legacy support. The Mac UI is designed with an "intuitive and consistent interface to the user" but what does this have to do with the hardware?
Constitutionally Correct
of course, what a nightmare that would be to configure for benchmarking.. i guess you'd have to use oracle on NT and on Linux, but it's a DOG on NT, and, as much as i hate windows, it just wouldn't be fair.
Use a better database program then. DB2 UDB is available for both platforms (bias alert: I work on DB2) and from what I see, DB2 runs pretty well on both platforms. That should even the playing field as far as database servers are concerned. There is no point in having one database vendor on one platform and a different database vendor product on another. DB2 is faster than SQL Server on NT anyway, so you'd be biasing the results before you started.
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
You assume that IBM, HP, Mindcraft and Dell are all in a big conspiracy to make Windows 2000 numbers look bad - are you kidding? The reality is that there is fierce competition for best SPECweb99 numbers, and Linux/TUX is just plain faster.
The other flaw in your argument is this TUX dynamic module. Check out the source code, TUX does dynamic modules. (besides, the SPECweb99 workload includes 30% dynamic load, so all SPECweb99 webservers must support dynamic applications.)
Ok here's the setup my co-worker and I used. We have 1 alpha server(DS 20) and two alpha workstations(XP1000) running True64 4.0F with all of their released patches. Also we have various linux boxes(PIII and dual Xeons) with the gigabit cards in them. We cat'd a file into /dev/null first to hopefully but it in the file system cache, then scp'd to /dev/null on the other machine. Also we did the same from /dev/zero to /dev/null. If you can tell me what software you used, I'd be more then willing to try it out since we need to get this fixed. Another stupid (unrelated?) problem we're having is it takes the cards about a minute to locate the network (maybe negotating) when they are loaded at bootup. This happens on both NT and linux so it's not the OS. If you have any ideas, please feel free to either e-mail me or post here.
RH have kept this one secret. If this is true it looks good!
http://digdug.cx/msdesign.html
--
Now the only advantage Win2K has over linux is a transparent start menu.
You haven't seen enlightenment window manager yet, have you ? Check out the EFM pages, and yes, it has had transparent menus for a while. But it also antialiases fonts and alpha mixes them for the transparency.
I pointed www.netcraft.com/whats at my own web server that runs NetWare 5.1, and it told me I ran IIS4 on FreeBSD. I thought nooit, that can't be right.
I've recently experienced this behaviour myself when I got another laptop, merely for demonstrating purposes. My machine at home is still a PII 266 with 100Mb internal memory. Windows 98 runs on an IDE (udma) harddisk and it runs decent. When I got my laptop (PIII 550, 64Mb internal memory) I've noticed nothing really special about the pre-installed Windows 98. It ran allmost as fast (or slow) as on my desktop computer. Some small details however did run smoother & faster (rendering in fireworks for example) but thats about it.
Linux on the other hand completely burns rubber. On my 486 server (web, print & file) dselect takes quite some time before I can enter the 's'elect screen. On my PII this time is reduced a lot but I still got to wait a few seconds. On my laptop its reduced to hardly one second (with the same amount of stuff installed as on my desktop).
This is just one, personal, experience but there are many people out there who can state the exact same experience with completely different hardware.
And this whole issue won't change as well and the reason is also very clear (and stated more then once, even on /. iirc). Linux is being programmed while Windows is "engineered" and as long as nothing changes these performance issues will remain. Where a Windows programmer is using a visual tool to produce some code for the "obvious" graphic routine's a Linux programmer is double checking if he made the right calls to the c library which he needs. Where a Linux programmer tries to squeeze the most out of the hardware by optimizing his code a Windows programmer uses tools which speculate on future, even faster, machines then the ones currently available. Small differences, big results.
Correct me if I'm wrong, but didn't the test of the M$ box have 1600 requested connections, while the Linux box received 4200?
I hate M$ as much as the next, but wouldn't this make a difference?
[1] What about stablity tests ? Put 2 servers against each other running non-stop for 1 month or so. After 2 days of getting hit real hard, I am pretty sure w2k would completly fall apart.
[2] Real like machines. Don't get me wrong, I know big companys spend $10,000+ for a server. But, what about a test aimed towards smaller companys. What about using AMD-500 - 700s, single CPU.
[3] Benchmarks != HTTP. What about FTP ? I have not seen this one used yet. (I could be wrong)
I know Linux would win hands down against w2k in any of these areas. It would still be intersting to see.
[4] I would like to see more dynamic content. Not so much ASP or PHP, but throw a monkey wrench in there. If they are using 20 clients connecting. Have 10 grab completly dynamic content and have the other 10 grab static HTML and some graphics. Then have them cycle through. There would need to be some pattern to it so the Windows folks wouldn't whine about it.
I think have a 2 server (with differant os's) running next to each other getting pounded 24/7 for about a month would be very intersting.
Why do all of the tweaking ?? Use a stock install. Use a few differant Linux distros just for shits and giggles.
until ( succeed ) try { again(); }
until (succeed) try { again(); }
Actually, they may have improved that a bit with 2K -- you can now run ASP connections in a new process, which of course Microsoft touts as a big stability feature.
I'm not commenting on testing methods exactly, but does it seem a little difficult to swallow that a Dell PIII/667 with one CPU running Linux beat a Compaq DS20 dual Alpha 667 running Tru64 by nearly 200 points?
A deep unwavering belief is a sure sign you're missing something...
You're on the right track but it'd make more sense to pick some algorithm that is actually useful to website development to do the testing with. Using FFT to benchmark a web server is almost as ridiculously as using static text.
It's modded at 2 because I have more than 25 Karma. I post at 2 by default.
Thank you.
--
Yes, and? Thems that live by the benchmark (Mindcraft and Microsoft) deserve to die by the benchmark. Let them eat crow.
James
The tcp/ip stack runs in kernel space. In the context of the kernel, there are no threads and there are no processes. Both of these are concepts that userland programs can rely upon because this same kernel imposes these virtual constructs upon them.
HTH. HAND.Pi
Ahh, that feels better. Thanks for that.
I'm not a Linux lover. I think it has a long long way to go before the mainstream starts to take it seriously. There are so many problems with it right now..installing programs, removing them, x windows interface complexity, simple text editors..the list goes on. Honestly, I don't think it will ever become mainstream - it will get replaced by something else that will before long.
You're not a Linux lover, which is enough to explain that you don't know what the fsck you're talking about. Installations in RedHat, for example, are fast, easy and reversible. Installations in Windoze are slow (how much longer do you think the redmond people can make the creation of a single 'shortcut' take?) and incredibly time consuming. How often have you seen the 'time to reboot!' prompt? I've upgraded 20 RedHat rpm's on a production machine in less than 2 minutes and did not require a reboot. Install Windoze and reboot 10 or 15 times just to patch the OS to a workable level. Install Office97 and its two monster patches and watch one hour crawl by.
I'm not a Linux zealot, but I do know that Microsoft is fucking evil. No question in my mind. Oh, and I'm a sysadmin with 15 years of Microsoft's bullshite operating system crawling up my ass so I know from where I speak. Linux may have a way to go in the consumer market, but the fact that Apache has over 60% of the web server market while IIS is losing ground should be enough for that weiner Billy Gates to be shitting his drawers.
BTW, this benchmark, as others have pointed out, is an independent benchmark and so you're comparisons with Mindcraft are `cat - > /dev/null`
:wq
10 LIST : REM MER : TSIL 01
Why are the two "benchmark configurations"
different at both setups.
w2k is only doing 1600 and linux
is doing 4200.
what does the fileset size mean?
cheerz
SmC
- In Memoriam: Jeroen de Bruin (1972-2004), bye bro
I read the disclaimer at the SPEC site. It seems the manufactureres each ran the tests themselves. This seems to mean that there was no common environment or proceedure to base these tests on.
I'd like to see Linux win this battle, but lets do it again on common ground with the same clients, same cables and switches, etc... Standardization, yea that's the ticket.
From the SPEC site disclaimer:
These are submissions by member companies and the contents of any SPEC reporting page are the submittor's responsibility. SPEC makes no warranties about the accuracy or veracity of this data. Please note that other results, those not appearing here and from non-member companies, are in circulation; by license agreement, these results must comply with SPEC run and reporting rules but SPEC does not warrant that they do.
Article X: The powers not delegated... by the Constitution...are reserved...to the people
Why specifically do you say Linux is not ready for mainstream? My experience shows exacly the opposite:
Oh, come on. Your post is well-written, but that doesn't make your arguments more true. For me, it's just one more unimportant piece of anti-linux advocacy, from the people which are annoyed by its rising.
Patola (Cláudio Sampaio) - Solvo IT
IBM CATE
SAIR GNU/Linux Certified
Patola (Claudio Sampaio)
Unix System Administrator
If I recall the Mindcraft period, people's "personal experience with Linux" consisted of "It kicks major ass on my P-120, therefore it must kick major ass on a 4-way Xeon. What? Oh shit. It doesnt!! Cheaters!!"
So ... what happens if, say, half of the Linux webservers switch to Tux over the next year or so? Do these webservers report as Tux servers, or Apache servers with a Tux kernel accelerator installed? The former could be a problem for purely stupid reasons: if 'Apache' held 30% market share, and 'Tux' held 30% market share, Microsoft would immediately claim victory in the 'web server war' -- as a result, all of the 'go with the market leader' people would begin installing IIS.
--
Tired of FB/Google censorship? Visit UNCENSORED!
G4 desktop machines OTOH were twitchy as all getout until OS 9.0.4 came out... now they are just a bit twitchy. And while we do have a beige G3 that crashes constantly (I still maintain that it's prob is a screw loose in the operator), we also have several that are solid as rocks. Mac's problems, in my experience, tend to lie in the OS not the hardware.
Insanity is the last line of defence for the master diplomat. But you have to lay the groundwork early.
Is that what Tux is?
If so, does that change things any? IIS is a system service, so it's going to have a lot more overhead than a webserver running in the kernel - but it's going to have a lot more protection around it as well. Benchmarking Tux vs Apache might be interesting.
- Steve
oops, wrong URL, the right link for the source code is here. It's standard user-space code, you can output any dynamic page with TUX as well.
GREAT, Thanks for the explanations! Now, is it possible to get more info about this TUX webserver? Is it open source? Is it available already? When kind of polling model does it use to share connections among threads? (sigqueues, poll(), something else??)
As far as I know, no details of "Tux" have been posted yet. The software availability is listed as "August 2000." However, it is definitely NOT stripped down for static pages. About 35% of the requests in SpecWeb99 are dynamic, including custom responses based on cookies, parsing and storing user registration results from POST requests, and doing real CGIs (must spawn a new process!).
--JRZ
Sure, more info is here. (which happens to be a comment in this thread :-) )
While I agree with you on this, the thing that makes linux either success or failure (in the enterprise sector) is it's performance and reliability. The boss doesn't care if you have fun with your system (perhaps he does, where are you working?-), instead, he wants it to WORK. In a big enterprise with huge loads on the server, speed is one considerable factor.
The freedom is the reason I use and develop linux. When good performance comes with it, I'm overwhelmed with joy.
I don't see this. While I agree with the general claim the thread-starter made (let's not let this benchmark have a free ride simply because "we" prefer Linux), I don't think the "let's give them the same equipment" type of test is the best indicator of real-world performance. Leaving other factors aside, what the speed-conscious webmaster wants to know is how much bang she can get for the buck.
Assume Mindcraft (third run around) is a "fair" result, in that both sides got the chance to tune their settings to their own satisfaction, etc. It would be a mistake to conclude, on the basis of those results, that one *ought* to use NT for one's website even if your only concern is speed (and that only with static content, etc.). Why? All Mindcraft showed was that NT beat Linux on a particular hardware configuration.
But that's not nearly enough information to go on in choosing a server platform (again, assuming one's only consideration is speed), unless you're choosing exactly the HW Mindcraft used.
No test on identical machines can determine which OS is "better" or (if you like) which you should use. This is not IROC, where every driver gets (supposedly) an identical vehicle; it's more like Formula 1 or Indy racing, where each team gets to put together its best combination of driver and car. To make the analogy even closer to that with choosing platforms, we could add the provision that every team gets $X to spend. That's the kind of metric you really want.
If one OS beats another (on whatever metric you care to measure) on most or all hardware that is comparable in terms of price, availability, and support, that's a valuable piece of knowledge.
"Oh, I hope he doesn't give us halyatchkies," said Heinrich.
If this is a hack, it could show what we all know is one of the strongest points of OpenSource: that you can hack (or pay somebody to do it) an application to suit your needs.
Yup, dude, its all about freedom
-- the computer doesn't want any beer, no matter how much you think it does. NEVER, EVER feed your computer beer.
Your problem is probably that your CPUs are spending a lot of time servicing interrupts from broadcasts. Putting a gigabit card in anything but a very high-end machine is probably going to adversely effect your performance, and you're unlikely to see much improvement in network throughput. Aggregating 100Mbs links (and running them duplex) is likely to give better results on SME*-level servers.
* - look Ma, a buzzword!
Well, you'd still do clusters for fail-over even if they weren't needed for scalability.
- Spryguy
- Spryguy
There are three kinds of people in this world: those that can count and those that can't
If I claim that I had a fight with Mike Tyson and he won, it's relatively unremarkable; the only implausible bit is that we might meet and fight in the first place, not that he wins. If I claim I had a fight with Mike Tyson and I won, such a claim is far less believable.
Thus, if your personal experience tells you that Linux kicks the shit out of MS operating systems for Web server performance, a benchmark test whose results accord with that experience is more believable than one which contradicts it.
That's just good sense, isn't it?
--
Xenu loves you!
do not confuse advocacy with information. FreeBSD and Linux are more or less at the same place when it comes to reliability, scalability, and network performance. at this particular point in time, I'd guess that Linux has the advantage with the improvements of the 2.4 kernel, but it doesn't really matter: FreeBSD and Linux are always catching up with each other; both teams are very good and neither will let the other OS get much better without getting better in the same (or equivalent) way. I'd say that, in choosing between Linux and BSD, you need to look specifically, either at personal preference and familiarity, or at the actuall support for the programs and services that you intend to run, and choose accordingly. Neither platform is overall significally better than the other.
Maybe it's just me being stupid, but how can the Linux box score 200% higher in generated traffic measured in ops/sec and only 6% higher in kbit/sec?
Does this indicate that the pages delivered were not identical?
> No it ain't. HW vendors like Linux because they no longer have to pay the "Microsoft Tax".
Somehow that doesn't keep Dell from being the biggest MS suckup in the whole business.
--
Sheesh, evil *and* a jerk. -- Jade
> There are so many problems with it [Linux] right now..installing programs, removing them...
Linux may have a long way to go, but installation problems certainly isn't any of the major problems. apt-get on Debian is insanely great. Try it out!
I would like to see a test of scale, Lets get the most powerfull and expensive whatever box that winNT/2K will run on and the biggest one Linux will run on and then compare them. Forget same hardware. Or another good test would be to make a budget. $10k build the best hardware configuration your OS will run on. Then lets see who has the price/performance race. based on hardware alone I am sure we could build a nice sparc or alpha that would kill anything window$ could run on.
Paying taxes to buy civilization is like paying a hooker to buy love.
The auto-negotiation protocol is a method whereby stations at either end of an ethernet link can communicate with each other using FLP (fast link pulses) and choose the fastest protocol that both are capable of using.
Auto-negotiation is relevant with UTP because of the multiple speeds that media is commonly used for. Everything from 10 megabit per second at half duplex, to 1000 megabit per second at full duplex is common.
Fiber optic lines are also used for all of these speeds, however 10mps cables use different connectors than those of 100 and 1000mps cables. If these cards are 100mbps only then autonegotiation just isn't relevant since there is only one speed to choose from and these cards are usually used in full duplex mode anyway because if offers twice the bandwidth.
Auto-negotiation on UTP most of the time boils down to choosing between 10 and 100 megabit (1000 megabit isn't very common yet) and half or full duplex. Full duplex means that the cards are sending and receiving at the same time, half means they are only doing one or the other at any given time.
The difference in media would not create any difference in performance since both UTP and fiber operate at the same rate and both are capable of full duplex operation. If both cards use the same chipset but only differ in the type of transciever used, which they would have to since UTP and Optic uses different transciever types, then there should be no difference at all in the performance offerd by either card. As far as a computer is concerned, the two are identical.
Lee
Muslim community leaders warn of backlash from tomorrow morning's terrorist attack.
Both Win2K and Linux were serving web pages from 4 striped drives. The difference is, NT had one drive for the OS and paging, and two drives for log files, whereas Linux had one drive for OS, paging, *and* log files. Seems to me Win2K had the superior setup.
(And the OS certainly does not have to "search" multiple drives or even multiple sections of one drive for a file. That's ludicrous.)
* And remember, it's spelled N-e-t-s-c-a-p-e, but it's pronounced "Mozilla."
telnet to port 25, I think they are a possible spam relay.
Only the State obtains its revenue by coercion. - Murray Rothbard
Congrats to everyone who helped make this day possible but the war has not been won yet... What will the next battle be?
--
Jake
Which begs the question: why, if the Mac's such a great box for servicg up web pages, are Apple running Netscape Enterprise on Solaris (see here)?
--
Cheers
Cheers
Jon
The benchmarks I think you're talking about are the SPECint and SPECfp '95 benchmarks. These were designed to test, respectively, the integer processing and floating point processing abilities of various processors. It was primarily a hardware test.
As such, manufacturers added "tricks" to their processors or compilers to essentially skip segments of the benchmark, getting highly skewed results with processors that weren't much faster. This is definitely a weakness of those two benchmarks. However, the description of the SPECweb99 test (linked to from main story) seems to take a lot of things into consideration. I don't think you can write off all SPEC products because a couple were faulty.
They always say, linux beat this, linux beat that. There are other OSes out ther you know, and they often perform better that Linux in server market. FreeBSD is one ;-) I am currently running Linux, but I don't want linux to be the common denominator cuz its a research platform where new technologies are being added every day, and some of them are not proven to be advantageous. Like adding a webserver into kernelspace for gods sake!
I though I will find at least one server from 'Linux friendly' IBM running Linux in the test. They supplied few servers with AIX, dozen servers with Windows, but none with Linux. Why?
It will be interesting to compare the comments of this article to the ones regarding mindcraft. I wouldn't be surprised if two thirds of the people here take this to be the truth simply because linux won.
Just looked at the specweb results. *Outstanding*
First of all, it gets about 90% scalability going from 1 to 4 CPUs.
Second, in the single-CPU configuration, TUX is faster than any other 2-CPU configuration (!). The 4-way TUX crushes all opposition - the closest thing (>20% slower) is an 8-way RS/6000 running the Zeus web server.
Lastime when it shows that Windows is the speediest, the icon used was the microsoft icon, which depicts Bill Gates as borg.
This time it shows that Linux is the speediest, instead of using the Linux icon, the Internet icon is used...
*sigh*
Uh. Read. It's both user space and kernel space, with the user space version being used for the specweb test.
It would be nice if someone would run some benchmarks with the same two machines, only have the W2k box serving up dynamically generated ASP and PHP pages, and the Linux box serving up a comparable PHP page. Whack up some identical code to perform Fast fourier transforms in the page and make it spit out the result. Once you get a database into the mix, you're also measuring the performance of it, and this is just a webserver test. Unless of course you have both boxes hit the exact same database, maybe a nice big Oracle database running on Linux. :)
Everyone here knows that MS zealots will say "Yeah, but W2k can spit out dynamic content faster...". It would be nice to have proof either way. I know I'm very interested in seeing how PHP on Linux compares to ASP on W2K.
Need Free Juniper/NetScreen Support? JuniperForum
I thought about talking about this, but I'd have to use my experience and that is fairly subjective. I'm curious what do you think the difference in human costs are? Lets see, take 30 minutes to install/configure Redhat, Take 10 minutes to install the latest updates, 5 minutes to disable services that are not needed and block the outside world from anything except 80 and 22, 10 minutes to install openssh, 5 minutes to load an existing website on the machine. Reboot the machine just to make sure and you're all set.
Looks like about an hour to me. Maybe an hour and a half if you want samba and frontpage extentions installed.
Just for kicks, lets take a look at NT.
When I installed it yesterday it took about the same about of time to install as redhat, so lets figure 30 minutes. Configure for network and reboot 5 minutes, Setup IIS 15 minutes, add webpage to IIS 5 minutes. Reboot the machine just to make sure.
Ok looks like a total of 55 minutes. Great, MS just saved you 5 or 35 minutes depending on what you're looking for. Is it really worth a few hundred dollars, if not more for an MS webserver if you really don't need one?
Also, with the linux box, I can ssh in and fix things remotely, I don't even have to be there to apply a patch when it comes out. As a consultant I find that very appealing. I just scp a file over, install it, restart the service and I'm set. NT I actually have to be there, when some of my clients are almost 2 hours away, I'd much prefer the linux method.
I find it interesting that there's no Macintoshes or Suns in the test, although there are at least one Alpha and two RS/6000s. How can they claim to be a useful benchmark by concentrating mostly on Intel hardware, and only running three HTTP servers? I'd think that the differences between different servers running on the same hardware could be just as much as between different hardware configurations; hell, even poorly configured vs. properly configured systems would be a huge difference...
Here I paraphrase a coughimaginarycough conversation over at Spec:
Uhh... what's this really big cheque doing here?
Which really big cheque?
This one here [hands it to coworker] with all the names on it.
[reading aloud] Pay to the order of SPEC... uhh... one million dollars... whoa... what's with all the names in the upper-left corner? Hah! Five signatures! Some joint account it was written on, I guess... B. Young, L. Torvalds, R. Stallman, L. Ellison, L. Augustin."
--
Personally, I want to see the BSD's go head to head. (as if they don't have enough rifts between them as-is) ;-)
Are you shure? It seems to me that Solaris has the TCP/IP stack threaded, and that is one of maior advantage for the Sun's OS.
Web administrators generally like to cluster machines with no more than four drives: any more and the OS spends more time searching for the file than delivering it. Assuming the network traffic they built up was the same in each test (again, they are a little shaky on that as well), Windows is taking more time to search across 7 drives vs. Linux's 5. Put 4 drives on each machine and you'd have a much better "real world" test.
While this isn't the total reason for the discrepancy, it's a good chunk of it. I'm also going on a hunch that the network traffic they created was similar but didn't really match up for both tests. We all know Linux is a good web server, but it's not that good. :)
- I don't care if they globalize against free speech. All my best free thoughts are done in my head.
What does this mean? Vendors obviously try to maximize # of connections, but they have to keep the bitrate above 320kbits to have a valid benchmark run. You can test with 1 million connections as well, but you'll get an invalid run because the kbit rate will be somewhere around 0.1kbits/sec. This is why you see almost identical kbits values (and all are a bit above 320kbits/sec), but different connections and ops/sec values. I hope this explains things.
See the SPEC-enforced Web99 Run Rules, there are alot of very strict requirements for a result to be accepted by SPEC.
The point is that it's hardly a good way to
run a benchmark when the machines arn't as
similar to each other as they can possibly be.
For every problem, there is at least one solution that is simple, neat, and wrong.
It seems as though RHAT has taken the trouble to render its TCP/IP stack into a multi-threaded model, rather than the forked model I understand it used to be. This was identified as the primary deficiency in the previous benchmarks.
Heh, lots of factors have been "identified" as the primary decifiency. Ask most people and they'll say it was the Linux SMP code. That was sort of the "popular explanation at the time", anyway.
There are other possibilities too. Later tests from c't magazine also showed Linux losing in tests when using two ethernet cards but winning the same tests when using one (Mindcraft I used four, and I think Mindcraft II was the same). This could also be the difference, especially considering how NT is supposed to have excellent I/O performance, but be rough on the CPU (with all those extra cycles used up for the GUI and the hardware abstraction layer, and so on). There's also the question of using an "off the shelf" Linux versus a high-end-specific version. Perhaps regular old RedHat would be more fairly compared with Windows 98/ME (in terms of both price and intended market) and RetHat's lesser-known "Enterprise Edition" should be compared to Windows NT/2000.
Your benchmark would be about as fair as the Mindcraft one.
--
Fuck the system? Nah, you might catch something.
>I've done on-site consulting work, and when you
>are sitting there on a customer site, both the
>NT and the Unix job will work out to 4-6 hours
>no matter what the little details are
ESPECIALLY when billing by the hour, huh?
-LjM
Well yeah, but, isn't that irrelavent?
- Spryguy
- Spryguy
There are three kinds of people in this world: those that can count and those that can't
Actually, in Windows NT, he's talking about the same architectural concept. Threads are pervasive in NT, in kernel mode and user mode. So he IS correct by saying that NT's TCP/IP stack is multi-threaded. In Linux, however, I understand that threading works differently.
HTTP Software: TUX 1.0
Is this some new httpd? What about apache performance? How stripped down is this httpd? Does it use the khttpd extensions available in the 2.3.x kernels?
Anyone know any info?
I'm guessing that TUX is a stripped down httpd for fast static page requests. Not quite a fully (mostly?) functional httpd like IIS. If this is the case, this comparison is invalid, comparing apples to oranges.
Using your sig line to advertise for friends is lame.
Um, 10 minutes to install OpenSSH?
You must really be slow.
Reminds me of a quote I once heard: "A child of age 5 could understand this ... Quick, fetch me a child of age 5!"
Seriously, if your kids can understand it, that's NOT a good indicator that other adults will understand it. I know -- I used to be that "6 year old whiz" myself about 18 years ago, and none of the adults even tried understand any of the stuff I was learning.
--Joe--
Program Intellivision!
I mean come on. If you don't use the exact same hardware then this benchmark is even more useless.
Course even if you do everything just right benchmarks are usually useless anyway. I have seen benchmarks that tried to prove that a z80 running cpm was better than w2k on a pIII. Ever notice that most benchmarks show how fast a computer can run games or work as a web server? Do computers have any other uses here people?
Then what you should do is test on the 2GB machine in both OSes, compare results, then test on the 4GB machine for the OSes that support it, showing how much performance is gained at that configuration.
An example would be, if computer A couldn't use AGP video with the current bios, but computer B could, you could benchmark both with a Voodoo 3 PCI card, then benchmark computer B with a Voodoo 3 AGP card, and say "We can't test directly, but we suspect computer A is missing out on n% performance by not properly supporting AGP." where n% is the difference on computer B. It still shows computer B kicking ass in the default config, but instead of making the whole machine look superior, it narrows the results down to the problem areas.
Its a good point, and my immediate response was that the Win2k results are so much worse that something must be very wrong.
However, if you go further down the results list, Mindcraft have also submitted a set of benchmark results which are broadly comparable to the Dell results on a different but comparable setup. It doesn't seem likely that both companies have made the same crippling mistake.
So it looks as though Red Hat hove done some serious magic with their threaded web server to me. Will they release the source I wonder?
How can anyone claim that any MS sponsored benchmark has any legitimacy whatsoever as long as MS insists that there be no benchmarking of their products in the EULAs?
I wonder how SPEC was able to perform this benchmark?
-Jordan Henderson
Sorry, but still no. He did not talk about a multithreaded TCP/IP stack, he talked about a 'forked' stack Linux had, which is a clear misconception. Even under Windows the concept of 'multi-threaded stack' and 'SMP-threaded stack'. Eg. it's possible to have a multithreaded TCP/IP stack on a single-processes system. ('multithreaded' in this context means: 'multiple threads can do socket operations' - there are OSs where only one process is allowed to execute TCP/IP operations, even if that process is not executing right now.)
How unfortunate for you. As you undoubtedly know, most 4 year college grads spent a large amount of time partying, not studying. Study was done simply to pass exams or write papers. Not to learn. And certainly very little of the education was about real-world situations and development timelines.
There are many people in the job market that are very qualified, with great credentials, but no degree. Many of these people will serve you better than most college grads, because they often feel they must prove their worth over the CS wizkid you just hired straight out of college. Unfortunately, the Wiz Kid you just hired out of college often feels an entitlement to a great job in the tech industry, without knowing a lick about the real world.
You can have the 4-year wizkids that learned 40 different drinking games and how to roll a joint, I'll at least take a look at those who don't have a degree but can prove their actual worth.
My experience, has been, with Win2k, unless you specifically change it, all files (even those with a .HTML or .HTM) extension gets run through the ASP parser before being served. The ASP engine is hideously slow. If in face the tests were on plain run of the mill html pages, this is a big waste! Ouch. Another bug in win2k.
But they still send question marks for quotes and ticks... :-)
--Joe--
Program Intellivision!
One of the outcomes of the Mindcraft saga was this wonderful set of benchmarks by C't.
One of the things that they did is force tests that stressed various parts of the OS. For me one of the more telling ones was the selection against many files, where the ability to serve off of disk (as opposed to out of RAM) was being pushed.
Linux won, of course. But I wonder whether Win2K is better at this than NT was...
Cheers,
Ben
My usual seat in the cluetrain is at A HREF="http://pub4.ezboard.com/biwethey.ht
actually I have, I run a 233Mhz computer. Transparency kills it. Plus, the start menu is provided by gnome (at least on my desktop), and as far as I know, it won't be transparent for a while. :)
"This is not a company that appears to be bothered by ethical boundaries."
Attorney General Mike Hatch on Microsoft
Don't tell me you benchmarked with ftp?
/dev/null
This with two acenics with default settings and ftp to
150 Opening BINARY mode data connection for test (1073741824 bytes).
226 Transfer complete.
1073741824 bytes received in 162 secs (6.5e+03 Kbytes/sec)
ok, let's tune them a bit (setting MTU to 9000), increase the socket buffer sizes to something sensible and also lets use a real tool
# description host sample_KB total_MB sample_KB/s avge_KB/s cpu_sec user_sec sys_sec sec/MB cpu_pct
1 source toy3 10500.000 745.500 71515.648 69882.607 0.120 0.000 0.120 0.011 82
(ok, probably the formatting looks like crap, but what that means is 70MB/s, not bad at all,
especially since one of the cards is an old obsolete model)
FTP uses silly buffer sizes (for gigabit speeds) and also you're getting the stuff off disk, which is probably slower than your network.
Tux the Linux Penguin is a mascot, not a piece of software.
If you are implying that Linux users jump on the bandwagon of everything even remotely Linux or OSS related, regardless of the quality of said item, there are numerous examples where you can be proven wrong.
Well, that is a bit strange yes, but what I did also see was that they don't seem to care much for security either, I think the slashdot peeps, should help them with there servers (ridicolous so many ports open):
... good.
W ME)
# nmap -nvvO www.andover.net
Starting nmap V. 2.52 by fyodor@insecure.org ( www.insecure.org/nmap/ )
No tcp,udp, or ICMP scantype specified, assuming vanilla tcp connect() scan. Use -sP if you really don't want to portscan (and just want to see what hosts are up).
Host (209.207.165.16) appears to be up
Initiating TCP connect() scan against (209.207.165.16)
Adding TCP port 32771 (state open).
Adding TCP port 4045 (state open).
Adding TCP port 80 (state open).
Adding TCP port 21 (state open).
Adding TCP port 873 (state open).
Adding TCP port 32773 (state open).
Adding TCP port 22 (state open).
Adding TCP port 25 (state open).
Adding TCP port 111 (state open).
Adding TCP port 32772 (state open).
The TCP connect scan took 11 seconds to scan 1520 ports.
For OSScan assuming that port 21 is open and port 1 is closed and neither are firewalled
Interesting ports on (209.207.165.16):
(The 1509 ports scanned but not shown below are in state: closed)
Port State Service
21/tcp open ftp
22/tcp open ssh
25/tcp open smtp
80/tcp open http
111/tcp open sunrpc
139/tcp filtered netbios-ssn
873/tcp open unknown
4045/tcp open lockd
32771/tcp open sometimes-rpc5
32772/tcp open sometimes-rpc7
32773/tcp open sometimes-rpc9
TCP Sequence Prediction: Class=random positive increments
Difficulty=286136 (Good luck!)
Sequence numbers: 7472F6AA 747E8F6E 748CD0F1 74931A18 7498F243 749AF9A2
Remote operating system guess: Solaris 2.6 - 2.7
OS Fingerprint:
TSeq(Class=RI%gcd=1%SI=45DB8)
T1(Resp=Y%DF=Y%W=FFF7%ACK=S++%Flags=AS%Ops=NNTN
T2(Resp=N)
T3(Resp=N)
T4(Resp=Y%DF=Y%W=0%ACK=O%Flags=R%Ops=)
T5(Resp=Y%DF=Y%W=0%ACK=S++%Flags=AR%Ops=)
T6(Resp=Y%DF=Y%W=0%ACK=O%Flags=R%Ops=)
T7(Resp=Y%DF=Y%W=0%ACK=S%Flags=AR%Ops=)
PU(Resp=N)
Nmap run completed -- 1 IP address (1 host up) scanned in 39 seconds
New things are always on the horizon
that these bytemarks always favor a perfect environment. In the real world, IMHO the platforms are neck and neck.
Go Mac OS X!
http://www.holymac.com/
- Carson www.holymac.com
We've seen Microsoft tout Win2k ASE without, it seems, having it available for download on their web site.
--
Violence is necessary, it is as American as cherry pie.
H. Rap Brown
(sorry for the messed up sentence, it should read: 'even under Windows the concept of a multithreaded and SMP-threaded TCP stack are different.')
Why does this sound vaprous? August is next month and since Dell is demoing it (and it is kicking Win2K ass) it must be very nearly ready for prime time. They are probably just ironing out some small problems so they don't pull a Microsoft and release a bug ridden product.
If you're going to hire a staff person (staff people) to administer your UNIX/Linux network, you will pay considerably more than for staff person/people to take care of that NT network. First, skilled UNIX admins are few and far between, and they require much more training so they can command more money. An NT admin can be had for much less, which is important when figuring out TCO.
this one says 4 (as in four)/
/ web99-20000626-00054.html
http://www.spec.org/osg/web99/results/res2000q2
the document itself says 2 (as in two)
http://www.spec.org/osg/web99/results/res2000q2
I am a linux/freebsd fan, but how precise are these benchmarks-results actually?
- In Memoriam: Jeroen de Bruin (1972-2004), bye bro
Ditto!
I think Microsoft needs to change the name of their flagship product to LOSE2000 = Linux Outperforms Superbly in the Enterprise :o)
No he meant FALL OVER!
I just wanted to refute slightly the above bit. My employer has just sent me a laptop obtained from Allboot.com (http://www.allboot.com/) which has Mandrake preinstalled. It's a PIII 650MHz,128MB RAM, big disk, 32 bit sound, etc. machine and runs flawlessly. It's the most beautiful PC I've ever used. So there. :^)
Oh, and no, I don't work for Allboot!
"We reject kings, presidents and voting. We believe in rough consensus and running code." Dave Clark, IETF
Microsoft heard that people were setting up clusters so that they could get fail-over.
Therefore Microsoft outdid themselves and made it a standard OS feature that it would fail over and over and over and...
Cheers,
Ben
My usual seat in the cluetrain is at A HREF="http://pub4.ezboard.com/biwethey.ht
When Linux didn't support 4GB of RAM, that was a liability. If Phil had a 2GB NT box and Joe had a 2GB Linux box, and they both needed more performance, Phil could pop in 2 more gigs and get it. Complaining about this particular point is like losing a tennis match and yelling "Hey, no fair, you've been working on your serve!".
--
Fuck the system? Nah, you might catch something.
and I think no SMP support in their TCP/IP stack), so I think it
would get roasted on this setup. FreeBSD's strengths are elsewhere.
One of the great advantages of open source is that one can have a
high-level of confidence that the OS doesn't cheat on benchmarks
(ie. by making changes to behaviour that increase benchmark
performance at the expense of overall performance). The temptation to
do so in a closed source environment must be pretty much irresistible.
Open source is definitely easier to tweak for performance. And easier to tweak for real-life workloads too.
FreeBSD is now backed by the BSDi people that have contributed better enhancements than anything SGI or IBM has put into Linux.
Show me any single machine that can run 40,000+ simultaneous instances of FreeBSD like IBM's S/390 can do with Linux.
Show me the LVM that BSDi has released for FreeBSD like IBM's LVM they've open-sourced for Linux.
Show me the journaling file systems BSDi has contributed to FreeBSD like IBM and SGI have done for Linux.
Show me the large memory patches, SMP patches, ccNUMA patches, and crash analysis patches that BSDi has contributed to FreeBSD like SGI has for Linux.
FreeBSD might very well trounce Linux in terms of stability and speed, I've not seen any numbers that show this one way or the other. But, I'd be very careful when making claims about what BSDi is doing for FreeBSD in comparison to what IBM and SGI are doing for Linux. Not that I'm saying that BSDi is not a good citizen of the open source community, only that SGI and IBM have certainly put their code where there mouths are in very far-reaching and irrevocable ways (there is no going back on code that has been GPL'd).
It would be interesting to see these benchmarks run twice, once with single-NIC servers and once with multiple-NIC servers. Windows NT2K aced the Mindcraft benchmarks by effectively coupling CPU's and NIC's. (One may wonder whether the Mindcraft tests were also run with single NIC machines, and the benchmark's sponsor [MS] elected not to have those results published.) Linux's IP stack has been continually improved since Mindcraft, so it would be nice to see how far the kernel developers have come since then.
--
Tired of FB/Google censorship? Visit UNCENSORED!
> Put 2 servers against each other running non-stop for 1 month or so.
That's the kind of thing I want to see!
Have both servers configured to automatically attack and respond to the other server's attempts to crack into / DOS the other server. When one server goes down and we see the "magic smoke" released from the hard drive, we know which OS is better.
"In this corner, wearing the Tux the Penguin case, and weighing in at 54 pounds, 3 ounces, the Linux/Intel server!"
(cheering of crowd)
"And, in this corner, wearing the Disintegrating Window (tm) case, also weighing in at 54 pounds, 3 ounces, the Win2K/Intel server!"
(cheering of crowd)
"Now, we don't want to see a fair fight. Do your best to cut the other machine off from it's services - FOREVER! Come out DDOSing!"
(bell rings to start the first round)
I'm sure we'd all pay for the pay-per-view to that one.
In post-9/11 America, the CIA interrogates YOU!
Is web benchmarking getting to be like 3D benchmarks where the OEM drivers get more and more optimized for benchmarking applications?
It seems to me that the purpose of benchmarking is to test the performance of some component. When that component becomes re-engineered purely to improve its benchmarks, has it gotten better at doing what it is supposed to do or has it merely gotten better at being benchmarked?
It's like SATs -- maybe as a generic "what have you learned" benchmark, they're not bad. Now you can go to school to learn how to master the SAT -- are you a better student or do you know more, or have you simply become a better at benchmarking?
I wonder if the TUX stuff really makes Linux a better webserving platform, especially considering that off-the-shelf hardware solutions coupled with reasonable web design can create a webserver which can easily deliver content equal to the upstream bandwidth available in most hosting environments.
All the PowerEdge servers I've bought in the last year (about 5) have used Adaptec SCSI hardware built onto the mobo. So though I don't know exactly what this "PERC2" controller is, I imagine it's Adaptec too.
Who was talking about setting up servers? Just to keep up the heat, why not look at this: E-Smith server-out-of-a-box. Apache, SMTP&POP3, DHCP, DNS, FTP, SQUID, web-based configuration & administration, dial-on-demand, masquerading and perhaps something more. It comes on a single cd which installs on your machine in MINUTES (so they tell me). Beat that, freak!
Admit it, linux is far more easily administered remotely than any Windows box. With ssh it's even secure. Not to mention the things you can do with little scripting with perl/php/whatever plus apache for instance. Web controlled ppp dialup, fetchmail retrieval, whatever you wish. Damned, you probably could build a web-based front end to build and install your KERNEL for that matter.
Now THAT'S freedom.
4) IIS SPECweb99 performance clearly suffers if more than one IIS thread per CPU is used.
5) you claim that Microsoft has no idea how to tune IIS - that is not a too credible claim IMO.
This is annoying. Why is it that everyone tests the highest-end hardware possible? I'd really love to see the performance of a 2-processor 2u server with less than 1GB RAM and processor speeds of less than 700MHz. Why? Because if I'm going to buy a rack of 20, that's what I'll be buying. I don't waste money on the bleeding edge when I can get more for less with stability.
Of course, in the MS world, you probably need 8GB RAM and 4 processors to run a Web server....
I wonder if the win2000 box really
had all 4 CPU's switched on.
Does anyone know how to check win2000
for SMP and if all CPU's are active?
Robert
First of all, these benchmarks (both the Win2k benchmarks and the Linux benchmarks) were posted by Dell, not by some random Linux zealots. Not only is that the case, but the other WinTel vendors have very similar scores for their WinTel hardware. Does this suddenly mean that all of the W2K vendors are conspiring to make Linux + TUX on Dell hardware look good? Or could it possibly mean that all research that Microsoft funded in the Mindcraft benchmarks is coming to fruition? My guess is that the folks at Microsoft are going to start to truly understand the power of release early, release often. While W2K has sat relatively still basking in its Mindcraft glory the Linux community has targetted the specific problems Linux had that caused it to do poorly in the Mindcraft benchmarks, and has rectified them.
Second of all this is a SPECweb benchmark. The web part of SPECweb would tend to indicate that it is a benchmark of http performance. If you read the spec you would notice that it specifically measures both static and dynamic http content serving. So while this does not necessarily mean that Linux is better than Windows 2000 it probably does mean that Linux + TUX is better than Windows 2000 + IIS (for the things measured by the benchmarks).
Your observation that most Internet facing sites don't have anywhere near this sort of bandwidth is certainly correct. However, my Intranet server does have this much bandwidth (not that I would appreciate it if it saturated this bandwidth). Besides, if you are going to let bandwidth be the limiting factor then it really doesn't matter what kind of web server you are using. A 486 running Apache will happily saturate a T1 with static content.
Not that any of this matters. The two most important features, to me anyway, of Linux are 1) Freedom, and 2) Cost. Linux wins hands-down if these are the factors that you value most.
From the results you must either conclude that Dell (and the rest of the WinTel vendors) are either trying to make Windows 2000 look bad, or you must conclude that Linux + TUX is going to make one heck of a compelling case as a web platform.
Either way it looks bad for Windows 2000 as a web server OS.
yes
I'd give my right arm to be ambidextrous...
i think you should have looked up what TUX was...
basically it's a webserver that works in kernel space.
this is something that you can do in linux and not in windows so it is almost a windows vs linux thing.
i agree with you that it would be interesting to compare apache vs TUX.
No, i dont think there is any such divide, and i think TUX does not contradict Unix concepts. CPUs get faster and protocols get more complex every day. Right now the HTTP protocol is common enough to be accelerated by kernel-space - just like the TCP/IP protocol got common enough 10-15 years ago to move into the kernel in many other OSs.
The question thus is not 'should we put HTTP into the kernel', but rather '*when* and *how* should we put HTTP into the kernel'. Think of this as an act of 'caching', the OS caches and should cache 'commonly used protocols'.
Where is the limit? There is no clear limit, but the limit is definitely being pushed outwards every day. HTTP is becoming a universal communication standard, with the emergence of XML the role of HTTP cannot be overhyped i think.
And the last but not least argument, if you dont need it, you can always turn CONFIG_TUX off.
Win2K Pro (for workstation use) requires at least 256MB of ram before it will handle 20+windows open at once w/o hitting the swapfile pretty hard.
I love to see open source software win in benchmark competitions against the big commercial heavyweights, but I have a hard time considering this a likely hardware configuration for a high performance webserver for use with either OS, however Win2K needs more memory to begin with, so the test really isn't likely to help sysadmins make the right choice.
Anyone can see through sloppy propaganda like this.
"Show me the money"
-Cuba Gooding Jr.
Amazing magic tricks
You have got to be kidding me. Apache on NT simply != Apache on Linux or on Solaris for that matter. Simply put, they are using the best http server for each OS, and the Linux OS (Redhat) won. I agree, the title should have been: Redhat beats Microsoft.
Make sure everyone's vote counts: Verified Voting
The media used should not be able to cause that much difference. So can someone tell me what auto-negotiation is and can it result in the huge difference in performance. If it does not, then in that particular benchmark, Linux performed better on its own merit and not because of some advantage offered by the hardware. Vivek Mittal
Vivek Mittal
Research Technologist
Telstra Research Labs
The reality is that most web servers are needed for stability and uptimes, and performance is second. The company I work at has both Linux and Solaris, as well I think that there may be some BSD here too. Solaris is stable and its hardware is also pretty good BUT not great. Most of the problems that we have had were because of hardware failures or defective hardware, not the OS. We also use multiple servers so clustering is a must and loadbalancing also.
Who puts a 4 procesor box on a web site? We do! We have many boxes that are 2 to 8 processor boxes. Of course we don't use that many intel boxes either except for Linux.
What makes this really funny are all the people who defend windows all the time probably not realizing that Yahoo, MS Hotmail, and yes slashdot all use UNIX or Linux. Yahoo uses Solaris and FreeBSD, Microsoft Hotmail does too. And guess what slashdot uses Linux / perl / and mysql. Hmm you visit a site that runs hardware you hate. hmm aren't you the hypocrite?
What windows 2000 really needs to prove is not that it can outperform LInux or Solaris, cause I am sure you can tweak it to be just as good if not better, it needs to prove that it can have 200+ days of uptimes on an extremely busy web site. So what company will be the first to have a large scale site and use win2k?
Lastly we recently bought a site that uses windows servers and we are moving them all to Solaris. Hmmm. Can we see some Solaris 4 processor boxes benchmarked against Linux and Windows? Oh we have and it blew both of them away!
send flames > /dev/null
Only 'flamers' flame!
b0rken
Hate stupid software on freshmeat? Laugh at
I just find myself wondering why BSD is not also tested when they line these things up? I am not a BSD activist mind you, but it would be cool to see how BSD ACTUALLY compares head to head with Linux, etc, ad nauseum... Peace, mOfOhAwK
Although I'm biased towards Linux, I'm wondering if this is fair (as usual). A few years ago, those nasty benchmarks that put Linux down were comparing a well-updated Windows system to a somewhat bloated RedHat system with a rather large footprint.
Now the tides have turned. Although everyone knows that Win2K is a resource hog (being buit over NT), how stripped down was the Linux system?
Don't get me wrong: I'm very happy to see the Linux systems beat the W2K systems, but I'm just pointing out that these benchmarks are never really comclusive because there are different grounds of comparison. It's like comparing an orange to a lemon, they're both citrus fruits, but although very similar in many ways, very different in others.
Just my CDN$0.02
Maybe that is as funny as:
;-)
http://www.netcraft.com/whats/? host=www.hotmail.com ???
I guess there are more people who don't trust Microsoft's Webservers
--
B10m
Why was there a three month difference (April/June) in the testing?
And how come the NT machine had a DELL disk controller while the RH machine had an Adaptec? You think they could have standardized on the Adaptec . . .
--
Never trust anyone over 90000.
SPECweb99 is the next-generation SPEC benchmark for evaluating the performance of World Wide Web Servers. As the successor to SPECweb96, SPECweb99 continues the SPEC tradition of giving Web users the most objective and representative benchmark for measuring a system's ability to act as a web server. In response to rapidly advancing Web technology, the SPECweb99 benchmark includes many sophisticated and state-of-the-art enhancements to meet the modern demands of Web users of today and tomorrow:
DELL ran the tests and sent in the results. Why whould DELL cook the books to make NT5 look bad?
.. just by having 10,000 slashdot readers click the story link all at once ;)
Thank you all that worked hard for making Linux what it is today ... thank you ... As for Micro$oft ... "watch out for the bird that can't fly..."
As Marty of LinuxToday puts it, though, "What does this mean? In the real world, probably not as much as it would seem. Benchmarks in general are typically set up in an ideal environment. Real world environments tend to be quite different. However, this does indicate that Linux is moving in the right direction."
I can't believe that people are sitting here saying "yeah, but this isn't the real world." Ok, no offense guys, I actually can believe it.
In Big-O notation, any scalar factor is neglible, it's factors like powers of the algorithm that arent, but this ain't an algorithm, this is a server.
If you haven't noticed, Win2K makes my computer at work crawl by relation to my computer at home... And my computer at work is MUCH faster. Trust me.
It beat Win 2k THREEFOLD. I don't care WHAT your real world situation is, THREEFOLD is a LOT. If it does THREEFOLD, that means that daggonit, it's probably going to be faster in "real world" situations too. Wake up and smell the coffee. Win 2K isn't the holy grail of computing. Linux isn't either, but it's serving 3 TIMES AS FAST, which is significant, unless the skew the benchmarks they were also running 50 copies of photoshop...
Eh...
If you examine the tests, you will notice that they are using the Alteon 180 switches for network connectivity in both tests, which are not available in 1000TX configs. They simply used different names for the same cards between the two tests (they used the 1000SX name under one of the NT tests also).
The servers in question are probably the exact same setup, the difference is that the PERC2 delivered the best performance for NT, but Linux couldn't support it (no driver). As such, they used the on-board SCSI that comes with the server, and used the MD raid software on the Linux side to do the raid. This means that the comparison is when using hardware raid on NT vs. software raid on Linux.
The difference between SCSI-2 vs. SCSI-3 is probably not an issue. In all likelyhood, they used the quad channel version of the PERC2 with the data striped across the disks on seperate channels, resulting in enough throughput to handle the job. In addition, the caching would reduce the disk load enough so the drive throughput would be less of an issue.
The final point is that Dell would want to demonstrate the highest performance they could, so would try to tune the tests to show the highest numbers they could. In addition, the numbers other vendors showed for NT on similar platforms was comprable, so probably the results are fairly reasonable.
Erik
Don't get me wrong, I like the idea of a Linux system kicking the sh*t out of W2K/IIS as much as the next geek, but it seems to me like this was a bit of a biased test. Really, it isn't fair to put a kernel space HTTP server which was built for the sole purpose of kicking @$$ on benchmarks up against a full-size web server thingie like IIS. I'm sure that if MS wanted to they could put out a server very simmilar to TUX and get comperable results. I'd like to see these benchmarks done again with Apache and see how they stack up.
I didn't once mention an MCSE. I said NT Admin. I am and NT admin (a pretty decent one, too) and I'm up at 2 AM reading Slashdot, drinking coke (well, caffinated beverages) and am quite happy at my job. When a server goes down, I know what to do most of the time without having to call MS tech support. Granted, my two NT webservers and 1 NT DB server haven't had any problems in the last year, but if something did happen to them I'm fairly confident I would know what to do. The UNIX admins at my company are definitely paid more than I am because they are difficult to find. A decent NT admin (again, not an MCSE paper person) who can really make NT do stuff is a little easier to find, and subsequently costs less. I have a little theory about this, though - maybe the UNIX people get paid more because the PHBs are scared of the command prompt? hmm..
Owners of such equipment should certainly thank Mindcraft. It was thanks to the kick in the pants that benchmark gave Linux folks that the appropriate changes were made to fix the problem.
We can still denounce Mindcraft as being a test that would be representative of real-world conditions to very few people (those who could afford a $ 50,000 server).
But in the end, it's good that kick was given - and we should congratulate everyone in the Linux community who worked hard to make those improvements possible.
D
----
I'm not sure where you get this, checking the numbers for NT, I get:
1 1598 99.9% 4408.4 363.5 2.76 330.5
2 1592 99.5% 4400.3 364.2 2.75 329.8
3 1598 99.9% 4409.6 363.4 2.76 330.3
and for Tux:
1 4200 100.0% 12291.1 341.0 2.93 349.7
2 4199 100.0% 12328.6 339.9 2.94 350.4
3 4200 100.0% 12309.0 340.4 2.93 349.9
These numbers are comprable in ratios, and match the pattern of all the other tests.
Erik
I've been using FreeBSD for many years, almost as long as Linux. I have a mix of BSD, Linux, IRIX, and Solaris boxes at my house. Three years ago, FreeBSD seemed slightly snappier on lower-end hardware than Linux, and both were certainly faster than the legacy UNIXes I was used to.
However, Linux still seemed to be slightly quicker on the high-end hardware of the time, especially since it had basic SMP support.
Things have changed since then. Blame it on all the money pouring into Linux, or all the kernel hackers Red Hat employs. I'm not sure what happened, but somewhere in there Linux got WAY faster than anything else, including FreeBSD. Today my current FreeBSD install seems absolutely sluggish compared to the latest implementations of Linux. (And I have FreeBSD running on faster hardware!) Granted, the difference is probably only in the range of 20%, but this difference is most definitely notable. And of course, as these benchmarks show, Linux scales quite nicely up to the super high-end. The filesystem, in particular, is blazingly fast on Linux as compared to anything else (including BSD).
Don't get me wrong: I love FreeBSD and I bow down in humble hommage to its authors. They have done an amazing job and given the world one of the best OS's it has ever seen. But the truth of the matter is Linux is 'better' by most standards, and the gap continues to widen due to the hype and the money behind it.
> Not very close, most busy sites don't have all static content.
You should understand the benchmark before you make blanket statements about it. SPECweb99 has 30% dynamic content (maybe fairly lightweight dynamic content, but it's a start).
See http://www.spec.org/osg/web99
I'm not rich. So sue me. Some of us can't always throw more fucking bandwidth and hardware at the problem. Especially those of us who make minimum wage because none of the godamn computer companies want to hire us because we don't have "degrees" or "credentials" or any of that bullshit.
Besides, our high school has a several thousand dollar tech budget and a T1 line and the shit still crashes every time you turn it on. That's no exaggeration, that's the literal truth.
I run linux on my laptop and it works. Best part it didn't cost me 10 billion fucking dollars which some of us apparently have to spend, it just took the physical cost of the CD plus a contribution to the free software ppl. 486 laptop does twice as much as any of the pentiums at school bleeding out a T1 do.
Ever get the impression that your life would make a good sitcom?
Ever follow this to its logical conclusion: that your life is a sitcom?
"I don't care about the Constitution!" --Bill O'Reilly, November 17, 2009
This article *was* posted on slashdot. Last year, when it was news. Now it's history. Next time look in the older articles before claiming nonsense.
moderators: this is a troll. Moderate accordingly.
___
___
If you think big enough, you'll never have to do it.
It's too bad they didn't benchmark exactly the same systems, with exactly the same RAID configuration. Dell's onboard hardware RAID has good Linux drivers on the PowerEdge 2400 and 2450, and this is a more reasonable system to be running a web server on anyhow - the 6450 is just overkill.
You can put together a pretty darn solid 2400 with Linux, dual power supplies, dual CPUs, and hardware RAID for something around $5000, which isn't bad. Seems like a good system for comparison, too.
Now before I get modded down, I be to remind whoever might read this that what I am saying is FACT. - bogaboga
Not very close, most busy sites don't have all static content.
. net
On a side note I think you should all visit this address and see what andover.net is running:
http://www.netcraft.com/whats/?host=www.andover
Solaris eh? Whats the front page of andover say?
"Leading the linux destination" great example you're setting there.
Only the State obtains its revenue by coercion. - Murray Rothbard
If you're going to hire a staff person (staff people) to administer your UNIX/Linux network, you will pay considerably more than for staff person/people to take care of that NT network. First, skilled UNIX admins are few and far between, and they require much more training so they can command more money. An NT admin can be had for much less, which is important when figuring out TCO
"You get what you pay for" When your network is down, and there isn't an explination in "the book" then what do you do? go hire another brainless NT Admin to say "You are going to have to be down 4 more hours while i use your company charge card to call microsoft tech support at 99.95/hr."
OR
Your unix admin who was working at 2am reading slashdot, drinking coke and being happy at his job (After all, he is a geek and geeks get paid to do what they love..) treats the problem as a new challange and goes after it and has it done before you know its down.
Once again, you get what you pay for.
This is truly fantastic news, for years linux has held the lead over Windows in stability, usability, remote access, and bugfixes. Now it's poised to take the lead in the one area in which it was lacking... meaningless benchmarks.
Now the only advantage Win2K has over linux is a transparent start menu.
"This is not a company that appears to be bothered by ethical boundaries."
Attorney General Mike Hatch on Microsoft
Are we that low? Do we pull a Mindcraft whenever we want?
These systems, although very similar, are not identical. Different drive arrangements, different scsi controllers.
And, to boot, one is running IIS 5 and one is running Tux 1.0 (whatever that is...).
What does this prove about the individual Operating systems? ABSOLUTELY NOTHING!
It shows that operating system 'A' running web software 'B' on machine 'C' is faster than operating system 'X' running web software 'Y' on machine 'Z'.
What the hell is 'Tux 1.0?' Yes.. I could look it up. WHy not at least benchmark Apache, so at least you could say 'benchmark of most common intenret platform for each OS' or something..
I'd use Linux if Windows was 200% faster..
A faster Windows still locks me into it's
stupid upgrade treadmill... Benchmark
results are just statistics.. and as you
know, there are "Lies, Damn Lies, and Statistics".
You can't just jump up and down when Linux
beats Windows on a benchmark. Then you're
setting yourself up to hang your head when
Linux loses one every now and then (Mindcraft)..
In so doing, you're missing the point:
The speed, usability, or even stability of
free software is not the driving force behind
its existence, It's the FREEDOM!
On Independence Day, of all days, you lose
sight of this? I'm so tired of these benchmarks.
-- The Funk, The Whole Funk, And Nothing But The Funk
Although unrealistic for the average business with an online site, there were a couple of configurations with only 1 gigabit ethernet card. With that configuration Linux served 1270 pages while Win 2000 served 732 pages.
Again, a 1 gigabit ethernet card is a bit overboard for most businesses but it isn't completely unreasonable. That is what you get when you ask for the vendors to put up what ever machine they want to have tested.
Anyway, if you look at the RS6000 hardware they had 8 gigabit etherenet cards.
Disclamer - Opinion of Person
To parafrase Mark Twain, there are lies, damn lies, and then there are benchmarks. Seriously though, what do they have to say now? It's especially interesing that Dell did both benchmarks. I would expect Dell to be Microsoft-biased. Especially since they, just like Mindcraft last year, used a system with 4 network cards. (for those who don't know, this is precisely why NT had won, and Mindcraft refused to do tests with only 1 network card).
___
___
If you think big enough, you'll never have to do it.
also, how do you figure that Macs are any less "desinged for this" than x86 boxes?
Because Macs--and yes, even your "quality-built" G4--are terrible at the most important factor for web-serving performance: memory bandwidth. "Apple's systems generally have had only about 60 to 70% of the effective memory bandwidth of contemporary x86 systems. This is due to Power Mac configurations that run the system bus at lower clock rates than comparable x86 PCs, and the simple fact that Apple's system ASICs cannot match the technical excellence of the best x86 chipsets like the 440BX." (source: Paul DeMone's Mac performance article at realworldtech.com)
Furthermore, as you'll learn if you read the rest of that article, Apple refuses to submit any Macs to any standard, fair benchmarking organizations, and in particular to SPEC, instead preferring to use decade-old discredited benchmarks incorrectly (BYTEmark) or make up their own. I wonder why?
the G4 is a damned powerful, quality built piece of equipment--better than most x86 boxes slapped together at some cheap ISP.
First off, it's OEM, but I'm sure that was just a typo. More serious is your perception that the OEM does anything which impacts the performance of the computer other than pick the components. The only thing that could possibly make a computer "quality-built" by an OEM would be making sure everything is screwed in tight. What matters is that the components themselves are quality-engineered. And in the case of chipsets--again, the most important part of a good web server--a plain old Intel 440BX knocks today's Mac chipsets silly.
And let's not even get into the x86 chipsets which are actually built to be used in web servers. Apple simply doesn't have anything to compete.
And there's no reason why they should. Apple has never ever pretended their boxes make good web servers. And considering all the things Apple has pretended over the years, that fact alone should clue you in that they probably don't.
Doesn't Dell own stock in RedHat??
.2c
Just curious... Would Dell be forming alliances and other things if they weren't interested in making RedHat look good??
Dell And RedHat Alliance
Yet Another Dell/RedHat Alliance
I'm not trying to start a flame war, just trying to see if other people see connections... This is yet another example of why benchmarks suck!!
My
The "Top 10" Reasons to procrastinate:
The "Top 10" Reasons to procrastinate:
10.
That's bandwidth vs latency .. the latency (amount of time spent by the computer processing the overhead of sending/receiving packets) slows down your copying to some maximum (3.6M/sec is pretty good on 100M LAN ... I did some file transfer tests (SMB) and got about 1.5M/sec from Win98 to Win98, and about 2.5M/sec copying from Win2K to Win98 (goes to show, I think, what a difference there actually is between Win9X and WinNT's networking ..).
Basically adding more bandwidth above 100 MB isn't going to increase the speed of your transfers much (you're probably sending bigger packets with 1000mbit fiber, whereas the max ethernet frame size on 100MB LAN is 1500bytes, so the overhead per packet should be less, so you should end up with faster transfers, but they won't be that much faster, since there is still more overhead on the hard disk, processing interrupts etc ..)
Mainly what you gain by increasing the bandwidth is that the network won't saturate as quickly .. i.e. with 100MB ethernet you might have 2 to 4 people copying files when it'll start to slow down .. but gigabit ethernet will allow more people to be doing 3.8M/sec at once (just as an example).
I have no problem with Linux/Unix. I've messed with it for about four years. I think it is great for what is built for (servers) and that is about it.
The EASIEST installation (Mandrake, I've found) is still almost impossible for the average person to use. I mean, unless you really luck out, it will take a long time for the average joe to be Linux-savvy. And this is what we are talking about, right? Linux being mainstream, right? I also work in a pretty technical workplace, and a lot of the employees haven't even heard of Linux! Sure, it's in Best Buy, and even some Wal-Mart's, but even by the farthest stretch of the word could you call the demand and usage mainstream.
Don't get me wrong, I'm only here for the best solution, for whatever I'm doing, but it's usually Microsoft. Bottom line, Microsoft provides the best tools for me right now, and I don't mind paying a price to achieve that productivity level.
They claim it takes a slight performance hit when doing so, but I haven't seen any benchmarks yet.
Even better, the whole thing is now released under GPL, which makes it a viable alternative to Postgres for the FSF fanatics.
The Web is like Usenet, but
the elephants are untrained.
Ok I know Tux is the penguin Linux mascot.. But I want to know what Red Hat Tux 1.0 is? I checked Red Hats web site and ftp site... nothing. Any ideas?
I mean sure, they did throughput test, CPU tests etc etc, but they were very calculated tests designed to test one thing only at a time (or something like that) and had little bearing on how a system/subsystem/software would perform in real life situations.
So the fact that Linux outperformed Win2k by a factor of 3 is pretty much useless as a comparison of real life performance.
Of course I could be wrong. I'm at work at the moment and can't get my hands on those dusty uni notes ... :-)
I'm absolutely loving reading all these comments right now. Its very amusing to me how all of you Linuz Zealots have jumped on this bandwagon proclaiming, "Linux is the greatest OS".
Yes, there were problems with the Mindcraft benchmarks - and yes there are problems with this one. Namely, what in gods name are they comparing? They certainly arent comparing operating systems - there are way too many differences in this case to do that objectively. Next time somebody runs benchmarks between the two OSes, please try to keep the following things in mind:
(1) USE THE SAME HARDWARE! I cannot stress this point enough. What you people may call minor differences may often have a MAJOR effect on the outcome of a benchmark such as this.
(2) Use the same Webserver Software. How in gods name can you blame or claim any of these benchmarks on the operating system? Both are using completely different HTTP servers (one which is isnt even publically available and shouldn't have been used = TUX). If you want a legitimate operating system benchmark and not and HTTP server benchmark - try to compare Win2k running Apache for NT and Linux running Apache. Otherwise climb off your high horses right now - these are webserver benchmarks NOT OS benchmarks. I for one will say that Apache for NT consumes a lot less memory than IIS 5.0 - though on my small intranet site I've yet to notice any speed difference.
(3) The results are unrealistic. What kind of server has 4 gigabytes of bandwith?
(4) Also - make a point to configure both servers equally - it seems to me you guys scrimped here and there on IIS configuration - I wonder why?
If the Linux world wants credibility - its time to grow up and earn it. You guys sure talk a great game - but when it comes down to the numbers you are either whining when you get trounced or creaming in your pants over benchmarks which are obviously flawed.
While Im on my soapbox - let me say this also: Its amazing how many of the news story's of this so-called "News for Nerds" site appear to be blantantly attacking Microsoft and promoting Linux. Its obvious that whatever sense of objectivity Slashdot once had (if ever) has long been lost to the horde of pre-pubescent teenagers who only have one goal: To get something for nothing.
So there you go - take it or leave it - I dont really care. You may either post a reply or email one to darkgamorck@home.com
Gam
I love idealists not because I am one, but because they make life bearable for pragmatists such as myself.
Suns, sure, but Macintoshes? I don't think I'm aware of anybody using Macs for even semi-serious webserving. Neither the OS (OSX is a different ballgame, granted) nor the hardware is designed for this kind of thing. Correct me if I'm wrong, please :)
As regards to the number of HTTP servers, maybe they just ran out of time and money to benchmark ten squillion different configuration, and chose the ones that they believed were in most common usage. Testing more of them would certainly be a good thing, though.
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
OK, I'm a linux zealot, but Be does really kick ass, especially on the desktop. My main problems with Be as a desktop platform is
A. Lack of applications....Linux just has more stuff, this is not to say that Be doesn't having anything, just most of it is graphics oriented, and I'm not a graphics person.
B. Lack of customizibility....Linux is simply more customizable.
For these reasons, I still use Linux as my desktop OS, though I still really respect Be and wish it well.
That being said, Be is far easier to setup and use, and, IMHO, tends to have better performance on desktop applications.
Unfortunetly, Gorilla aside, I think Be's switch of focus to Internet Appliances (ala Stinger) will end up dooming the OS (dispite Be's claim otherwise). Afterall, what company is going to want to develop an application for an OS that is not even the primary focus of its mother company. And open source people are likely to stick with platforms that are completely open source.
On a side note, I'm not optimistic about Be's chances as an applicance OS either. Linux is also competing for this space, and its free. In devices that are only supposed to cost a few hundred dollars, this is a big advantage over even a resonably prices OS like Be.
Yup. It's flame bait for Linux (although I'm not sure I see why it's flame-bait for micro$oft)
In your item 1, you state "it's very difficut to configure". That, of course, depends on what you want it to do. My experience with Linux has been basically that, if you can follow directions, you can install software. I'll admit it's not as easy as clicking 'Next' in a nice Wizard, but there's been nothing I couldn't get working with the right directions. As for 'make install', there are several package managers out there (from RH, Debian, primarily). My personal experience is with RPM, and I find it every bit as easy as the "Add-Remove software" in the Control Panel.
Your second point is that micro$oft is more commonly 'pre-installed'. I won't deny that, but as for the vendors that DO support Linux (such as Dell), when you're configuring a Dell Server, be sure to look at the 'OS' section item. Linux is there with all the other NT and Novell options.
From your comments, it seems as though you don't distinguish between End-User computing and Server computing. Personnally, I feel that Microsoft has done their job well in you, because that's exactly what they try to do. They try to convince people that GUI is more important to them then the functionality/stability of an OS. They want people to feel that wizards, icons, menus, and pointers are more important than having a robust server. They want feature-rich clients talking over proprietary protocols to proprietary servers to drive a companies infrastructure, and let's face it.. they're EXPERTS at achieving that.
I, on the other hand, still can't understand why I want the overhead of running a graphics interface on a system that, in an ideal world, sits in a room, in the dark, by itself, hopefully, untouched. It bothers me to even INSTALL an X-server on a UNIX Server, let alone run the server @ Run Level 5 (that's the typical level that brings up the XDM login window, for those who don't delve that deeply into the OS.. An IBM S390 is an OUTSTANDING hardware platform that is scaleable, robust, and fault-tolerant. It has been for YEARS. And yet I'd never put one on a users desk (or should I say I'd never put a users desk on one...)
Linux/Unix's strong point has never been it's desktops. It's the stability reliability, and effeciency of the OS that makes it good.
UNIX GUI's have come a long way, and I can't WAIT for the day when I can enjoy UNIX stability in a desktop platform, with all the software and peripherials that the Micro$soft users enjoy today.
-- You can't idiot-proof anything, because they're always coming out with better idiots.
I realise that some guy from LinuxWorld effectively made the same comment as me and it is in the original news post - oops. I wrote this post in response to reading a few reader posts about how great Linux is cos it beat Win2k...
IIRC, the admin who posted that article dismissed it as FUD from Microsoft. I don't see anything similar in this case.
In addition, your suggestion doesn't even make sense. If I thought that what I had to say was nonsense, then why would I bother posting it all? And I think it's highly impractical to sort through each slashdot article before posting to an article.
I also find it funny that you're asking moderators to moderate your post as a troll.
moderators: the parent post is flamebait. Moderate accordingly.
Please note that SpecWeb99 is highly software dependent, however. A lot of the test centers around two things: TCP/IP and network card performance, and software implementation of the dynamic part of the test suite.
In general, Windows 2000 can beat Linux performance in the TCP/IP stack at the high end. (According to my own tests at >300Mbit). Linux tends to cap at certain speeds. We're looking into why this happens but will be feeding results as we get them to kernel development lists.
The software implementation of the dynamic suites is also highly important in the tests. There are easily 50x differences in a CGI implementation vs. the IIS implementation that is included with the suite. Presumably, a more efficient implementation is also possible. This likely also had something to do with the score difference.
Lastly, Linux is a much better webserver platform that Win2K. This is not necessarily due to performance, but to stability/management/etc.
Spec is a funny test, and pushes webservers in ways I don't think a good benchmark should. It emphasizes TCP/IP performance too heavily IMO, and depends too much on a vendor writing a good dynamic suite. While this does give flexibility, it also mitigates the actual OS choice in SpecWeb99 score.
Anyway... great news for Linux, and I'll be looking forward to getting my hands on more specific information about this test. It is nice to have a benchmark go our way every once in awhile...
---
Drew Streib
---
Drew Streib, dtype.org
hmm, business is not influenced by some sort of philosophy. If you want to get Linux into the business world you need some "real" arguments. No many in the business world give a something about that their software is free. You have to get some arguments like "ROI" and "TCO" to convince a suit to deploy Linux instead of windy products. This arguments are simply numbers.... And to calculate em you have to benchmark.... Sad but true, but welcome to the real world!
The two major distinctions between these benchmarks and the unjustly-maligned Mindcraft benchmark that were later confirmed by PC Labs:
1) these tests compare Win2k to Linux. By contrast, the Mindcraft study compared WinNT4.0 to Linux.
2) in the "Operating System" column of the Linux boxes, we see a revealing note:
Operating System: Red Hat Linux 6.2 Threaded Web Server Add-On
It seems as though RHAT has taken the trouble to render its TCP/IP stack into a multi-threaded model, rather than the forked model I understand it used to be. This was identified as the primary deficiency in the previous benchmarks.
At the time, Linux afficianados claimed that the superiority would be short lived. Assuming these stats are otherwise legit, it seems as though they were right, and in such a brief period of time as well. I'm impressed! Keep pumping out impressive turn-arounds like this one, and very soon commercial entities will have to give open source its just props as a development model.
I am slightly curious whether this "web server add-on" is available to consumers, and also whether it is a fully-featured web server. If not, and this is just a hack, that might cast a pall of illegitimacy. Anyone have the inside scoop?
-konstant
Yes! We are all individuals! I'm not!
-konstant
Yes! We are all individuals! I'm not!
Notice this is a speed-with-certain-capabilities test. Other real world factors such as stability are not tested -- if your server crashed during a test, it's not a valid SpecWeb test and does not count at all. See the rules, where several requirements are defined which a crashed system will violate.
Any protocol that relies on TCP is going to dissapoint further, as the smallest latency will introduce some sort of rate-limiting effect. Big bandwidth is about multiple streams at once, not about how fast you can download .. leave speeds above 100Mbps to routers and switches for now.
If you ask me, common sense would dictate that any sort of "benchmark" that claims Linux to be (apparently) the best, would automaticly be popular. Frankly, I find it hard to belive that a well-established OS on it's native hardware could be beat out by an Intel/Linux combo.
I'm not suggesting that Win2k is better than Linux (as that I'll easily believe), but perhaps someone got carried away when they fabricated this, and decided that an RS/6000 on AiX (64-bit remember)couldn't beat Intel hardware.
I swaer by OBSD and AiX, and I've found both to be mis-represented.
This article is from Oct last year so it is a little dated. If you assume that any of it was ever true (other than the defenition) then it is quite amazing how far Linux has come in a few short months./ LinuxMyths.asp
http://www.microsoft.com/NTServer/nts/news/msnw
-- Hail Eris
Even though I like linux, and use it extensively....
there are lies, damned lies, and benchmarks.
See you, space cowboy...
I know the parent post is a troll, but just for the sake of clarity here are some urls, including the original /. story.
The more you know, the less you understand.
1) the maximum filesize in the SPECweb99 benchmark is 900kb, this is why there is a 1MB limit set. Your claim that there are 1MB objects in the benchmark is false.
2) the CGI executable is mandated by the SPECweb99 Run Rules. A process must be created and destroyed. But the total amount of CGI requests is 0.1%! All the other 99.9% of the workload was handled with IIS 'low application priority' modules, which is a DLL loaded into IIS's address space, not a .EXE.
3) the IIS object cache was set to 2GB (not 2MB). It's set to 2GB because Windows 2000 + IIS has a serious limitation, threads (such as the IIS threads) can only address 2GB. This is a design flaw in Windows 2000, which hunts them in the enterprise now.
4) are you really seriously promoting the idea that the top 4 PC OEMs (Dell, IBM, Compaq, HP) and Microsoft did not tune IIS to the max and somehow conspired in making Linux+TUX numbers look good?
Fact is, the only reason why the TUX result was compared to the same Dell system is that the Dell system also happened to have the fastest Windows 2000 results. Your whole line of argumentation is obviously flawed if you compare IBM's similar Windows 2000 SPECweb99 result to the TUX result.
Actually it's funny you should mention Microsoft's Linux Myths page. Those reports are posted on Slashdot.
As far as credibility goes, I guess it's you that's lacking. Linux zealot bashers are worse than the Linux zealots they're bashing.
Enjoy your job, make lots of money, work within the law. Choose any two.
Quoting one email from Alan Cox is as representative of linux kernel development as using only one verse from the Bible to sum up all of Christianity.
Begone troll!
As I remember the ZD benchmark comparing NT & Linux last year did not do that well, remember, I hope they will run a second one this year... running TUX as HTTP software and beat the crap out of NT.
Disclaimer: "These opinions are my own, though for a small fee they be yours too"
--- Bouh !!! ---
I've stopped caring about linux now. I think Open source is a great thing. I like CLI's, so naturally I like the unix idea. But until something major changes, I don't think Linux will take over from microsoft in the consumer-arena. Why you ask? Because
1) Its very difficult to configure. I have very mainstream hardware. Nothing funky on the motherboard, 3com NIC, graphics card from Diamond, SB soundcard, etc. But I could never get everything to work at once. Keep in mind I'm fairly computer literate (I 've only built my own computers since I was in 8th grade.... ) and I know what I'm doing. But I could never get everything to work together, and this is with three different distros keep in mind. If someone like myself, who knows about computers, can't get the damn thing working what makes people thing that average joe-consumer and idiot-boss will be able to make it work on their computer? And that's not even getting into installing software ("make install" my ass, there's better ways to do things if you want it made easy). "Well, just buy it pre-installed then!" you might say, which brings up
2) A Micro$oft OS is pre-installed on almost every computer on the planet. "But dell has linux preinstalled on some laptops!" If they do, they're not making it very visible, a search a couple weeks ago in their home-user laptop section turned up nothing with linux. ditto for small-business section. Which leaves the average joe to install it himself. Refer to 1) for the impossibilities of that happening.
Now, you're probably wondering why I said "Be zealots" up there, right? Well, that's my solution. I think with the right pushing that Be actually has a chance against the gorilla. Unfortunetely, it dosen't look like that's going to happen, it looks like another OS/2 (that was a fun one to play with, btw). Coulda, shoulda, but didn't because of piss-poor advertising. Make no mistake, I think Be is great, I use it as often as I can. Its easier than anything to setup (just install it) and it works great out of the box. I urge everyone to try it. Hell, its even free. And parts of it have been open sourced.
My rant is done. I guess I could sum it up by saying " We're screwed, the good stuff always gets squashed by the gorilla ". Have a nice day :)
Predictions for the moderation: Troll, Offtopic, Flamebait. Lets see how close I get.
Linux is only Free if your time is worth Nothing
Linux is only free if your time is of no value
Be in Your Senses
I'm sure each vendor did everything they possibly could to improve their SPECWeb99 results, since it's in their best interests. Does this mean that Linux is just better overall? Does it mean that it can be twisted the most to win any benchmark if you try hard enough?
The notion you get from reading linux-kernel is that they're totally against patching the kernel just to win a specific benchmark, but it obviously did very well in this one.
Then again, there are lies, damn lies, AND BENCHMARKS. I see this as being more credible than the Mindcraft benchmarks (Mindcraft, haha, that sounds suspicious) since it wasn't simply NT vs. Linux and multiple vendors are involved.
It would have been interesting to see FreeBSD thrown in, just because it's another open source system. Maybe there's a trend here? Easier to tweak open source systems to win benchmarks? Maybe they're just clearly better? Hmm.
"What does this mean? In the real world, probably not as much as it would seem."
Interesting. The Auto-Negotion would have a possible affect if the hub/switch cannot support a 1G/sec transfer rate, because it would negoiate either a 100M/sec or a 10M/sec connection based on what the hub/switch could handle
what is Really interesting however is the fact that the Linux system gets the 1G/s Fibre Network and the Win2k gets the CAT 5. I was always told that the Fibre networks are better because their is less crossover and more reliable connections overall. My question is that why didn't they release these tests using the same Idenital hardware, Especially when the hardware is critial to the benchmark they are trying to obtain.
The other thing I noticed about this was that the controller cards and hard drives were different as well, although this could have little to no factor to the test, I'd feel a little more confortable with these tests if they used the same machine and hardware.
In Soviet Russia, Trojan exploits YOU!
I don't think it is fair to say that Linux beat Win2K but that the web software( TUX 1.0 ) beat Microsoft's (Internet Information Server 5.0) If you wanted to say Linux was better, than you would use the same HTTP software for each system. Although this may have it's drawbacks, depending on which type of file system and such each system uses.
Standard I/O Error. Incompetent/Operator.
Take a look at the actual lists of tested OSes. They certainly could have tested more! :) Settle more than one eternal argument!
Anyhow, for the list:
1.AIX 4.3.3
2.AIX 4.3.3 + APAR IY09807
3.AIX 4.3.3 + IY06844
4.Compaq Tru64 UNIX V4.0F
5.HP-UX 11.ACE
6.Microsoft Windows 2000 Advanced Server
7.Red Hat Linux 6.2 Threaded Web Server Add-On
8.RedHat Linux 6.1
9.Windows 2000 Advanced Server
10.Windows 2000 Server
Three AIXes and not one Solaris or BSD. Sheesh!
-bugg
The problem here is that these figures aren't aimed at freedom lovers - most of us take benchmarks with a pillar of salt anyhow. They are aimed at business users. Most business users don't play politics - they want results, and they want comments from the press they can show to their managers.
So while the benchmarks don't directly impact on us, their influence over business computing does give benchmarks some significance.
tangent - art and creation are a higher purpose
postmoderncore - art and creation are a higher purpose
Wrong. Onlye fiels named .asp is parsed through the ASP engine. The thing MS improved is if you call your files .asp and only have plain html in them, they will run faster than before. But if they just call the files .htm then they will NOT go through the ASP engine.
Actually it's funny you should mention Microsoft's Linux Myths page. Those reports are posted on Slashdot.
See here.
As far as credibility goes, I guess it's you that's lacking. Linux zealot bashers are worse than the Linux zealots they're bashing.
Perhaps you're saying that because you're a linux zealot?
Network Security: It always comes down to a big guy with a gun.
There should never be a 3x difference on similar hardware between two operating systems which are similarly tuned. Despite each side claiming how much their OS is superior (Mindcraft and then this) there should be some obvious reasons for the discrepancy. Looking closer at the config:
Linux has net.tux.max_cached_filesize = 100000000
Win2k has Inetinfo/Parameters/MaxCachedFileSize=1000000
IIS was only allowed to cache in memory 1M files while TUX could cache 100M files. Possible difference here.
Linux has net.tux.max_backlog = 3000
Win2k has Inetinfo/Parameters/ListenBacklog=1000
IIS cannot have as large a backlog as Linux. Probably not as big an impact.
Linux has 1 disk for OS and logs, 4 disk software RAID0 stripe, using 2MB chunk size, for fileset
Win2k has One 9GB 10KRPM disk for OS and paging, 2 for logs, and 4 striped for web pages
Note that software RAID/Mirroring on Win2k for logging is madness - it will kill performance. Linux can log to a single drive. Probably not a great performance impact but hard to tell without performance monitoring.
Linux uses Onboard Adaptec AIC-7899 SCSI
Win2k uses Dell PERC2
What is the difference here? Perhaps this is significant, but I'm not sure. The disk I/O is very suspicious as the killer and I'm wondering if the PERC2 drivers on Win2k need lots of work. Be interesting to see some performance stats on these.
Ok. I'm trying to defend NT a little and that's suicide on Slashdot. All I can say is despite the collective view here that Linux is better that Win2k, I find it hard to believe that anyone thinks a 3x difference on the same hardware is what is expected.
John Wiltshire
Fear: When you see B8 00 4C CD 21 and know what it means
These numbers seem hugely high to me. I mean... 4,200 simultaneous connections at 350kbit/sec is around 1.5Gbit/s. To do that you'd need some fairly serious NIC's. A closer inspection of the test setup reveals the server was pushing 4 networks through 4 gigabit alteon network cards.
:)
Reality check guys. Does anyone have 4 gig of external connectivity? And doesn't 4,200 simultaneous connections of 350kbit/sec each represent, like, Yahoo? (without doing the sums)
This would also seem to spurn a more serious debate in terms of web performance testing. If we can get a single server to munge through this kind of quantity of throughput - why have clusters of servers at all? Clearly real world servers perform nothing like as well as this, and we need to have a better look at why.
Dave
I write a blog now, you should be afraid.
Considering it out performs Win200's webserver with http://www.cs.helsinki.fi/u/mjrauhal/linux/cola.ar chive/1997-07/cola.1997-07-1 3.010
I've got a 66Mhz 486 running GNU/Linux, 450 days uptime serving up to 10 000 hits a day...
Danny.
I have written over 900 book reviews
Slashdot as pseudo-intellegent cache anyone?
I write a blog now, you should be afraid.
ok, you're wrong. Mac OS X Server has been out for well over a year, and it does a handy job of serving up web pages with Apache. also Web Objects, from the NeXT world is a nice piece of software for delivering web-based applications.
more info can be found here.
also, how do you figure that Macs are any less "desinged for this" than x86 boxes? the G4 is a damned powerful, quality built piece of equipment--better than most x86 boxes slapped together at some cheap ISP. sure it may not be a high-end Sun box, but there's no reason a Macintosh can't serve web pages with the right software (and better than an x86).
-j
I believe that he someone at Spec did the study and showed it to his superiors and that they were too brain-dead to notice. I am, in fact, so gullible that I believe that the people at Spec then released it to the press, and in the face of all the inevitable reaction, have not reviewed their findings.
I also believe that slugs have wings.
Pull the other leg, it's got bells on it.
To mail me, remove the 'mailno' from my email addy.
"Yeah. It smells, too..."
Don't be surprised. NT has no fork function and its CreateProcess API call is extremely slow. MS moved IIS to threads very early on primarily because starting new processes on NT is so darn slow. Apache got bitten by the same bug/feature. When Apache was ported to NT the first time around, it was dog slow because was acting like a good unix program: it forked. Later Apache-win32s used threads. There's jsut no way to get good performance on NT iif you need to make new processes. NT is a VMS derivitive, not a unix derivitive, and this is where it really shows.
OK.
First, in CmdrTaco's own words, the article has "Some good points. Some not-so-good points." It doesn't say "look at this FUD nonsense from Microsoft."
Perhaps you're saying that because you're a linux zealot?
No, I'm saying it because you suggested that those types of reports never appear on Slashdot, when the very same report you used as an example was indeed posted.
Enjoy your job, make lots of money, work within the law. Choose any two.
You really read the article "indepth", didn't you?
Upon seeing the box was too small, Schrodinger's Elephant breathed a sigh of relief.
Suns, sure, but Macintoshes? I don't think I'm aware of anybody using Macs for even semi-serious webserving. Neither the OS (OSX is a different ballgame, granted) nor the hardware is designed for this kind of thing. Correct me if I'm wrong, please :)
You mean besides the military. A G4 running OS X Server is nothing to sneeze at, and if memory serves correctly, web serving on the mac using WebObjects is a pretty sweet combo... If you are going to include commodity x86's then you should include macs...
course if you're running linux PPC, then you can run Apache on a G4 and really rock and roll......
~ a low user id is no indication I have a clue what I'm talking about.
There are counter-examples, sure, but the examples far outweight those.
Don't get me wrong, I've been using Unix for five years and Linux for three and I love both of them, but there's a certain mindless mindset that too many Linux users partake in.
"News for Nerds, Stuff that Matters" definitely needs to be changed to that.
It is a shame that this is what slashdot has become. Zealots out of one corner, trolls out of the other. Something becomes truth because it matches their beliefs.
Moderators are moderating based on this belief, M2 doesn't seem to be doing much better . . . We're getting stories submitted with misleading information or outright lies.
Everyone has to start taking a step back and looking objectively at all the facts. This isn't what we, a community that is supposed to be open to other views and encourage free expression of different thought, are supposed to be doing. It's almost as if /. has been turned upside down lately. There are accusations of the editors censoring comments and there are even those who would have this be. COME ON! Has it been this long since we fought against mandated censorware?
I'm not saying that what the Trolls said actually happened, but this is what is going on in our community right now. Perhaps the allusion that follows is brought on because I just finished watching the movie again, but this is too much like The Matrix. We have people who, like Cipher, are willing to give up all our rights--something that we would cause us to take up arms if it were IRL--for the pleasure that is bliss.
Changing your threshold doesn't do enough when moderators make biased decisions. This is a note to all the moderators out there: don't you get tired of reading the same stuff every time? "Linux is good, Linux is great, I surrender my will as of this date." Come on. For the rest of us, do something useful. Moderate up those posts that have a novel idea, not one that just says something you agree with.
And for everyone else: RTFA. Read the articles before posting. A (-1, Redundant) is a pain, especially since it is impossible to read every post, but when you state something clearly stated in the article it wastes everyone's time. If the site the article is on is being slashdotted, wait until you read it before posting. As I said before, this is for everyone, including moderators. You have your moderator points for a while, you have time to check validity of posts before moderating them up or down.
And what happened to moderation totals anyway? I like knowing what's happened to a post.
It looks like the dates the tests were run were a few months appart. If I go to Dell today and say I want a PowerEdge 6400/700 with a SCSI card they're going to ship me that model with whatever SCSI card they can get right now. If I come back 6 months latter and order the same model there's nothing that says they are using the same SCSI card, they might be able to get card X cheaper so they use it. I've noticed the selections change on Dell's desktops over time, why should servers be any different?
Well, benchmarks mean nothing special to begin with, but Apple would be ill-served to submit for this benchmark, BECAUSE 1. Apple flip-flopped so badly on processors for a while that Motorola concentrated on a multiple-market (read "embedded" PowerPC603) architecture and is now well behind the curve for power/speed, even though the G4 at 500 MHz performs better than a Xeon III at 733 and between an Athlon 700 and an Athlon 750. 2. The Apple bus and peripheral speeds are way down compared to PC FSBs. Catch-up for a proprietary design is harder. Apple has much going for it and is a wonderful platform for many things, including web services, but their rather small market is not well-served by advertising differences in speed (which used to change hands frequently until Intel and AMD really lit the afterburners competing for market share), Besides, Apple is still trying to live down the marketing fiasco about the (nonexistent) prohibition of exports of the G4 because it was too fast. Please don't get me wrong, I am a fan of MAC. But I find it unsurprising that there was no G4 here. Now in a competition for aesthetics, or for cool and economical operation, Apple takes it hands down. (Course, the little NEC Powermate might win "cute")
First, in CmdrTaco's own words, the article has "Some good points. Some not-so-good points." It doesn't say "look at this FUD nonsense from Microsoft."
I never claimed that he said "look at this FUD nonsense from Microsoft." I said he dismissed it as FUD. Look at what he just said in that quote. "Some not-so-good points." I'd say that supports my statement, wouldn't you?
No, I'm saying it because you suggested that those types of reports never appear on Slashdot, when the very same report you used as an example was indeed posted.
Or maybe it's because slashdot is composed of a bunch of sheep who can't think for themselves, the majority of which blindly supports Linux and hates Microsoft.
i'm curious about this tux 1.0, too.
according to the bencmark results, it threads, so it's not apache. (pre 2.0) it says it's not available till august of 2000?
what the hell IS this web server? it looks more like AOL server or Domino (god forbid) than anything. It certainly is not apache.
OK, now I'm pissed off - you can't even buy a dell 6400 with linux on it, TUX doesn't exist.. Anyone have an answer, here?
there is no mention of it ANYWHERE on dell's site, and nothing relevant on red hat's.
so, here's my guess: you'll be seeing a product announcemnt for Red Hat's new web server here on slashdot in a month, and a lot of pissed off apache developers, since this is exactly the direction apache is heading.
unless these crazy bastards are using apache 2.0 alpha, and whipping 2k's ass left and right with it. that would be superbly funny.
--
you know you want soccerchix, dammit
i browse at -1 because they're funnier than you are.
I agree with you - MAC OS X seems, from what I've read, to be a decent server OS. I thought I'd acknowledged that already.
I also agree that Macintoshes tend to be well-built machines. However, all the ones I've seen are desktop machines, not servers. There are x86 boxes out there expressly designed and built as servers. While not everyone uses them, if I was doing important webserving they are the sort of thing I'd certainly want over *any* desktop box. I agree that while an average desktop PC from a typical x86 vendor is no more suited to webserving than an arbitrary Mac, you can buy ones that are.
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
Comment removed based on user account deletion
So why not make sure Linux wins a benchmark in an area where they know Linux is popular already? Fudged numbers and fudged hardware aside (we'll assume they were honest), it would certainly seem a logical thing for them to do. As is evident from the results page, they blew away the competition with TUX 1.0. I've been unable to find any information on this (please enlighten), but it appears that it's a kernel patch, because the options are set with the kernel interface. Is this the khttpd that was discussed after the Mindcraft fiasco?
In any case, if it's in kernelspace, it's most likely not a full-featured HTTP server like Apache, Zeus, IIS. So it can spit out static pages as fast as you'd possibly need. Big deal. Fireworks accelerate faster than space shuttles, but you wouldn't create dynamic content with fireworks. (Erm... where was I?)
My point is, Dell has proven that a specially designed static page server is faster than servers with more features. That doesn't really tell us anything we didn't know. It doesn't demonstrate that one OS is better than another, nor does it make deployment decisions any easier (except for fools).
Gates' Law: Every 18 months, the speed of software halves.