If someone tries to convince, say, your boss that you should use NT instead of Linux for your website (say you're hosted by a partial T3 and only serve static web pages) and cites these results, will you say: A) "You're right. These benchmarks prove that NT is superior to Linux and we'd be foolish to go with Linux now, though if we work real hard coding at Linux et al. linux will be able to beat NT at benchmarks so we can switch then."
or B) "Actually, if you look at the tests rather than the soundbyte about them, you will see that they prove that a Linux/Apache system will be waiting on our internet connection anyhow, and Linux is a much better value protosition than NT is, remote administration to a box with no mouse or monitor is extremely easy, remote displaying comes standard (no need to buy citrix), you get greater stability, and source code that uses open standards so that if life takes one of its many unpredictable turns and Linux is no longer the best solution, we won't be trapped into expensive solutions that we don't want to be in due to vendor lock-in."
So do we give up, or work with what we've got? I suggest that everyone who's whining about "let's stop whining and get back to coding" at least say what you mean, "Let's just give up and try to beat microsoft on their own game instead of doing what we want." The people who do the coding are still doing the coding. This is about advocacy and marketing. I suggest that the people who do Linux Advocacy/marketing don't stop simply because everything didn't go perfectly. These numbers are only a defeat by microsoft if we let them be. Remember, there is more than one part to the Linux community, just as there is more than one part to the body. Just as I wouldn't suggest that those who do documentation stop doing documentation and start coding, I also wouldn't suggest that those who do marketing stop and start coding. It takes all sorts to run this world, and we shouldn't start neglecting any part of it, marketing included. Microsoft is going to try to put as much spin on these tests as they can. If we stand by and do nothing, we'll be as guilty as they are of the spinning. Sins of omission are still sins.
I don't suggest that we do it defensively, though. I suggest that we do it confidently and aggressively. act like we're in control. That's the neat thing about self-confidence. If you act like you know what you're talking about, people tend to believe you. So why should we give up acting like we know what we're talking about.
We've got some test results which prove that Linux can handle the needs of 99% of the world. Why exactly should we hush that up in favor of letting it be thought that Linux is slow.
It's so nice of Microsoft to pay for this apache advertising. Just as a point of reference, 1800 hits/sec is the same as 155,520,000 hits/day. I think that it's safe to say that noone in the world gets more than 150 million hits per day of static content. Wait, there's a better way:
1800 hits/sec * average 2k/hit * 8192 kbits/kbyte = 29,491,200 bits/sec, or 29.5 MBits/sec. What's that now, a T3 line? I know that a T1 line is 1.5 MBits/sec. Ok, so apache on one of these boxes can fille the equivalent of 19.6 T1 lines by itself. If (a bit more realistically, how many 2k files get those types of hits) those are just 10k files (let's not get into pictures), that's 147.5 MBits/sec, more than filling a T3 line, IIRC, and definitely filling up aapr. 98.3 T1 lines.
What's the problem with Linux/Apache, now?
May I suggest, if you can afford this sort of bandwidth, that you buy one of those 32CPU sun E10000 servers and call it a day? (or a server farm of linux boxes, since you're serving up static files.)
Oh, if you're serving up >1800 files per second of 2k files, who are you?
Oh, one more thing. If this is all on an intranet, you'll still need Gigabit ethernet if you're serving up the 10k+ files, so the sun box still applies to you.
Do even believe what you're saying? Do you really mean that Linux should be able to beat NT at utterly contrived situations that have little to no bearing on real life? Can you point me to one sight where people are requiring thousands of pages per second of static content for only six minutes?
I don't particularly care about samba/NT fileshare performance. That whole network neighborhood is so damn slow on the client end that a slow server wouldn't really be noticed. That thing is a pain to deal with when there are ten compuers on the network. I couldn't imagine how slow it would be with 100. Or the thousands that this test is trying to simulate. Anyhow, I don't really know much about file sharing, nor do I care much about file server performance. Other people can worry about that.
What part of this mindcraft test has any bearing on reality? It's a benchmark contest, pure and simple. It's an NT machine tuned for this benchmark versus what was a Linux machine tuned for bad performance on this benchmark. Now maybe it will be an NT machine tuned for this benchmark versus a Linux machine tuned for this benchmark.
To quote penguin computing's FAQ, "Now, this is a complicated question, and one that benchmarks can't really answer (unless of course, your intended application of the computer is to run benchmarks, in which case benchmarks will be a great indication of the performance you will get).
To quote someone else, "Disraeli was pretty close: actually, there are Lies, Damn lies, Statistics, Benchmarks, and Delivery dates." (found in fortune.)
Benchmarks are virtually useless, especially highly tuned benchmarks. What's the point in saying that if two OS's are configured in such a way as to achieve optimal speed for a useless application one can do that useless operation faster than another.
Static pages server to as many people as are being simulated using hardware with beta drivers has nothing to do with the real world. I defy you to come up with one example that has the sort of usuage pattern that was tested in the mindcraft test (i.e. >2000 page views per second of static pages for only 6 minutes).
So in a sense you are right, you can't dictate the external conditions for a server environment when making a server, one builds the server to those conditions. That's exactly what didn't happen here, the conditions were build for the NT server.
I congratulate the windows people that they can make an OS which can do so well on pointless benchmarks against crippled products. That's really something to be prowd of. Now pardon me while I go challenge my grandmother to a game of basketball...
It's good to know that in a corporate environment one only serves static web pages for 6 minutes at a time. And that one picks hardware that one knows is only supported by beta drivers. And that one doesn't tell the people who are trying to help you to optimize your system what's really on it, or, heaven forbid, let them actually see the box. That's what really happens in a corporate environment.
And no corporate environment ever moves over to NT on the client side, nor do they use win98 clients. They immediately reformat their hard drives and install win95 on all those new boxes that they get. Thanks for that wonderful description of what happens in the "corporate environment".
Don't ask me why I fed the troll. I really don't know. I just can't let that ever-so-slightly credible argument go unanswered.
How about a 6 hour test? That would certainly disable the ability of programs to hide performance hits over short periods of time. Does anyone really care how many hits their webserver can take in 6 minutes and then stop serving?
I remember hearing something about NT being configured to do all logging after the test, not during, while apache was configured to log during the test (with host names -> DNS lookup).
Not to mention, the chances of NT failing under a really heavy load in 6 hours is much greater than in 6 minutes, really helping things out to be more accurate. Noone runs their webserver for 6 minutes.
Remember that glide is being written by Daryll Struass (sp?) in his spare time without being paid. The voodoo people are basically using Daryll. Remember that in this case the "customer" is being given the drivers by Daryll, he's got no reason to besides the goodness of his heart.
If you want someone to whom you are the customer to write drivers, ask the makers of the voodoo chipsets to write their own drivers and do it on a timely basis for a funded company. Daryll's doing an amazing job considering that he's not being paid for it.
Please note that app->video card only works in the few instances when the app is simply writing pixel data (such as image viewers, etc.) and doesn't need any functionality from the card aside from being able to display passed pixels.
Most programs like things such as acceleration from video cards, at which point no sane programmer wants app->video card, that's just painful as hell as then you have to start supporting the acceleration of all different video cards, which is what X is around for in the first place.
So while what you're talking about is useful for unnacelerated video playback, it doesn't have much applicability to anything else. Wouldn't be that bad an idea, though, except that it would be a bit hard to work out the security model for.
It's called the MIT shared memory extension. Most programs that do intensive graphics stuff (TV players, the gimp, etc.) already use it. X and the program share a section of memory. The program writes to it, and then notifies X when it's done. Then X simply uses that area of memory for moving things to the card. The kernel isn't really involved at all except for setting up the shared memory area and for the pipe protocol to notify the X server. This also allows the X server to properly do clipping, and things like that.
I know that C2 doesn't mean much, but could you publish publicly this info that you have a bunch of standard slackware Linux boxes that have a C2 rating? It would be nice publicity for Linux, especially for those who have no idea what any of C2 security means.
The guy addressed this question earlier. They've got an old version of SunOS/Solaris which isn't Y2K compliant. They have to upgrade, and the Solaris licenses are quite expensive.
That's what someone said earlier, I disclaim any knowledge of knowing whether or not it's true.
On a different note, Sun boxes come with a real paucity of software, as far as I can tell. O fcourse, most of my experience is with older sun boxes, not newer ones, so I don't know if they do include all the utilities that I'm used to with Linux (such as compiler for more than 5 languages, make, bash, perl, etc.)
Note: I'm not advocating blindly replacing everything with Linux. It'll take a while before Linux is ready to be put on 64 CPU sun boxes (though I hear that there are people working on High end SMP stuff).
Actually, how water does tend to freeze faster than cold water, at least in some places. Hot water allows certain gasses to escape more quickly than cold water (I think that chlorine is one of the main ones) and thus can freeze at a higher temperature. At least that's what my earth science teacher tought me back in the eigth grade. I don't know if it's true or not, but something about certain saturation diagrams I saw in chemistry seem to point to that conclusion.
Just a side note.
Oh, btw, Xterminals were and are a good idea, but they're virtually unavailable nowadays. Besides, can you get any that have great acceleration for 1024x768 at 24 or 32 bpp and have 10/100 support for full duplex ethernet?
NFS is viable on trusted networks when you are root on the main computer. We use it in our computer lab, and it works quite well. On the other hand, I don't want to have to set up nfs exports to edit a computers/etc/hosts file, or something like that. And remote display works especially well for debugging X apps that work on your computer but not someone else's. As well, when you just need a quick and dirty solution to get something done without having the time to set up NFS and then take it down again, remote display is quite useful.
As well, remote display is good for thin clients, so that you can have a good graphics card and a cheap CPU.
On other big benefit, is that you can run other people's binaries. For example, I'm posting this comment from my Linux/Alpha box, in a netscape which is remotely displaying from my comp. sci. server (Dual PPro 200 Linux box). Quite convenient, really.
So yes, NFS can be advantageous in some circumstances, but remote display can be quite advantageous in others. It largely depends on the circumstances and what you want to do.
There is currently a driver under development for the Matrox G200 cards. It's pre-alpha right now, but development is active and progressing (I'm hoping to contribute once my G200 PCI card shows up in the mail).
Do you have plans to support glx modules such as the current (development) glx module for the matrox G200 cards under Linux, and when it comes out (july-ish, I think) Precision Insight's glx DRI? It strikes me as strange that Q3 will require a 3DFX card since it's written in openGL? Btw, as 3D accel. for Linux gets better, such as the G200 driver, will you release AXP (Linux/Alpha) binaries, since we'll have 3D hardware acceleration as well? Thanks.
You might want to check out Cygnus's GNUPro utilities, which have lots of PII/PIII optimizations, I think. THey're fully GPL'd, too (not free, though, they cost $).
Do you run your NT boxes on NT certified hardware? I've seen rooms full of NT boxes, and they were semi-constantly being "rebuilt". In many people's experience, NT (and other Microsoft software) has a habit of comiting suicide. It has way too much of a reputation for being suicidal to ignore that.
I don't think that it's as much NT's kernel as NT's structure - being a black box (with hazy error messages), there's not much that you can do as a user to fix things. Linux (and UNIX in general) is very open, so when some library isn't working, you fix/replace the library, not the whole OS. When an NT library isn't working properly, your usual recourse is reboot/reinstall.
I'm not saying that there aren't some exceptions, but NT, especially to one is isn't a guru, isn't nearly as fixable as a UNIX box is (especially a Linux box).
Anyhow, this isn't meant as a mindless NT bashing session, but just poiting out that NT does have some problems in this area. About 50% of the NT boxes in a bank that I worked for had a habit of comiting suicide on a semi-regular basis. Note: this isn't as much those NT boxes running only on-CD stuff, but those running third party apps (like citrix, watermark, and securID). NT does seem to be fairly stable if you haven't installed anything that comes doesn't come on the NT CD.
The primary domain controller was really funny. It would continually keep eating up memory until it exhaused memory, then it would die. And it was a dual PPro 200 with 256 MB RAM. Pretty funny to watch, actually.
It's not the average Linux advocate, it's the average loud mouthed slashdot ranter who wants everything for free. Plenty of people, myself included, have paid for RedHat, and plan to continue to do so. They make a decent dist, and they give a lot back to the community.
It is nice that RedHat has their dist up for free, btw, for things like the comp. sci. department of my school where we're strapped for money and there's a semi-lengthy process involved in actually paying for anything, so for the computers in our lab we just download the RPMS and install.
But I (for personal use and at work), and many other people pay for RedHat. We're happy to. RedHat deserves the money.
Please don't confuse the loudest people with the most people. There's a world of difference, as the fact that RedHat is still in business and doing well proves. If it was only the minority of people who didn't want everything for free, how do you think that RedHat would still be financially alive?
1. In their scowering of the net, did they send mail to Ask Slashdot? I don't recall seeing an Ask Slashdot about configuring Apache optimally.
2. They should have posted their httpd.conf file, not given a few random tidbits from it. It looks suspiciously like they had apache configured as an inetd service, not running normally. The only way to tell, of course, is for them to post that file (and the other config file).
3. Why didn't they do the tests with the default parameters? Since they claimed that they didn't know how to configure apache, they should have tried the out-of-the-box config to see if they just screwed up. The Apache performance tuning page says that tampering with MinSpareServers and MaxSpareServers is unnecessary. They set them to 1 and 290, respectively. And they set maxclients to 290. I would like to see some actual records, like the output of: `while true; do ps axm >>/tmp/ps.txt; sleep 10; done`. It looks like they either started swapping or as some other posters indicated ran out of file descriptors.
4. Did they turn on apache's cache? They seemed to turn on the IIS cache.
What I'd really like to see is the results if they had just used a standard apache config, perhaps upping the MaxClients only, since they're planning on using more than 150 simultaneous connections, but they should have tried it without even messing with that.
Also, did they restart the apache server *after* the reconfigured it? They didn't document their actual test process. They also claimed to have tried different apache configurations. What were they? What was the performance on them?
Maybe Linux/Apache needs improvement, and maybe not. But this test contains way too little information to be fit for anything more than the recommendation that "more testing is needed". Did they even run more than one test for their results? Did they run a whole bunch of tests and just report the one that they want?
This test doesn't include nearly enough information for proper evaluation, and it isn't all that meaningful. I don't doubt their conclusion, that on the particular machines that they used with the particular configurations that they used they found that certain products on NT were faster than certain other products on Linux.
The real question is how representative were their tests? Can anyone really represent the information that they gave as being enough to tell? If not, that test doesn't merit anything more than a "that's interesting".
Does anyone know how transferable Linux knowledge is to UNIX? Of course, it depends on the UNIX that one's talking about, I know. I assume that Linux knowledge will transfer to a sys V based system better than to a BSD based system, since most Linux dists are based on sys V standards (correct me if I'm wrong).
Essentially, how well does being a Linux sysadmin prepare you for being a solaris sysadmin, or an AIX sysadmin, etc?
> As for the slavery, it is completely irrelevant, > and there are so many things about that > paragraph that are irrelevant that I am not > going to bother to respond.
Interesting cop out. My question was pointed to determine your ideas of social morality, etc. There are is a small group of individuals would would say that the slavery should be legal, after all, the guy took out the morgage knowing what would happen if he got laid off. It's his fault for not developing enough skills to cope in a difficult world, etc. From what you've said, you seem to be one of them.
I want to know if you're one of those people who basically ascribes to the doctrine that the world doesn't owe anyone a living, so you shouldn't take care of anyone else. Note: I'm not attacking you personally. You're probably a kind, generous person. But the views that you expressed sound like the views of the dog-eat-dog proponent, that is, one who believes that because a dog can eat another dog, and sometimes will, that's a perfectly fine situation.
Most of us believe that some work should be done to make the world a better place. Sure, some people will steal if you let them, so you do your best not to let them.
You see, the problem with microsoft is not that they don't have the "right" to do what they've done. In the Hobbsian sense they have the right to do anything that they can get away with. On the other hand, they don't have the right to do what they've done in the sense of deserving it, or it being morally permissible. Just about all of the incompatabilities between M$ products and the rest of the world are artificial. They're designed into the system to act like a drug - to hook you and make you dependent on them.
Which makes me think of a side note - do you think that narcotics should be regulated? Do you think that crack dealers are simply "good businesmen" (they are selling a product for which there is a large demand)?
Anyhow, the argument against microsoft is that they have no right to compete artificially. Just like it should be illegal for microsoft to compete by filing hundreds of lawsuits against its competitors, or sending them hundreds of pounds of mail every day, or flooding their email box with messages, or ping flooding their competitors websites into oblivion, etc. It should be illegal for them to build addiction devices into their products.
But the problem with us coming to an understanding seems to be that your view of morality is "anything not directly violent is ok". Which works, in a sense. Of course, I can't see how, using that premise, you could support things like the civil rights movement, etc. Why should the government dictate that blacks have to be able to use the same bathrooms in private establishments as whites. Or why should blacks be able to sit in the front of the bus like whites in privately owned buses. Or why shouldn't a company be able to hire only white males solely because they're white males? It's their business if they want to do that.
I'm assuming that you're against equal employment laws, correct? And you believe that child labor should be acceptable because either the kid or his parents should be responsible for the kid. If he wants some extra pocket money, why should someone stop him. And I assume that loansharking should be perfectly legal. And that electricity and water companies should be able to charge whatever they want.
If the water company decided one day that they were going to charge everyone 100% of their assets for a drop of water, that's fine.
And there's nothing wrong with what OPEC did, is there? And I assume that price-fixing should be perfectly legal.
Out of curiosity, since you're dependent on your parents for food etc. in the first X years of your life, do they in effect own you? Should they be able to starve you, etc. if they want? Send you out naked into the snow if they don't want you in their house or wearing their clothes?
What do you propose to restrict? What laws do you say should be on the books?
There are lots of monopolies, many of which are strictly controlled. Look at your electric company (depending on where you live). Look at your water company. They're often monopolies, and often very highly regulated. It's not restricted to microsoft, nor was it invented by microsoft. The general principle is that if you can exert undue influence over your market, you're not going to be allowed to.
As far as Microsoft being required to publish specs, there is precident for that, as well. Ever go to the supermarket? Every food item which isn't composed of one thing only (i.e. bananas, apples, slabs of beef) are required, by law, to include both caloric information and the ingredients, listed in order of volume. Shouldn't companies be allowed to put anything they want into the food that they sell and not have to tell us?
Of course not. Why? Because then they place people in danger. If you don't know what you're dealing with, you might take something that you're allergic to. Why should M$ be forced to publish the full version of all their specs? Because without them, they can exert an undue influence on their market. If I write a story in Microsoft Word, I should be able to retrieve it, even if I no longer have Microsoft word. They have no rights to my intellectual property, and no right to keep it from me. By using an undocumented API, they are, in effect, keeping it from me.
Can you come up with a convincing argument for why selling something doesn't inherently imply selling the information on how to use it and interact with it? Can you give a convincing reason why a company should be allowed to make a product which hides my information, or the information that someone else gives to me, in a format that only they can read so that I am forced to pay them for it (without invoking the morality that if I steal from my neighbor but he can't force me to give it back, I should be allowed to keep what I've stolen)?
Think of it this way: do you think that it would be OK if Honda sells you a car, that can only use Honda ExpensoTires, and they're the sole manufacturer of them? And if you only found out about it afterwards (how many software boxes say "This product uses a proprietary format that binds you to our products and doesn't allow interoperability with other people, excepting that they also use this product")?
Out of curiosity, do you believe that slavery should be legal, if, say, people became slaves by selling their freedom rather than being cought or sold by someone else? E.g., Henry takes out a morgage. Henry is then laid off from work, and can't pay his morgage, so the bank reposeses his house. Unfortunately, property has devalued, and the house is only worth 80% of the morgage (and let's not forget interest). So the bank sells henry to pay for his debts. Do you think that this should be legal or illegal. If you think that this should be illegal, how do you reconsile this with your statement that microsoft should be able to do anything that they want, including product dumping, market control through proprietary APIs, etc.
Slashdot Mirror
Thanks for posting this. It's awesome to get some real life data to throw into this. I wish that more people would do this. Thank you.
If someone tries to convince, say, your boss that you should use NT instead of Linux for your website (say you're hosted by a partial T3 and only serve static web pages) and cites these results, will you say:
A) "You're right. These benchmarks prove that NT is superior to Linux and we'd be foolish to go with Linux now, though if we work real hard coding at Linux et al. linux will be able to beat NT at benchmarks so we can switch then."
or
B) "Actually, if you look at the tests rather than the soundbyte about them, you will see that they prove that a Linux/Apache system will be waiting on our internet connection anyhow, and Linux is a much better value protosition than NT is, remote administration to a box with no mouse or monitor is extremely easy, remote displaying comes standard (no need to buy citrix), you get greater stability, and source code that uses open standards so that if life takes one of its many unpredictable turns and Linux is no longer the best solution, we won't be trapped into expensive solutions that we don't want to be in due to vendor lock-in."
So do we give up, or work with what we've got? I suggest that everyone who's whining about "let's stop whining and get back to coding" at least say what you mean, "Let's just give up and try to beat microsoft on their own game instead of doing what we want." The people who do the coding are still doing the coding. This is about advocacy and marketing. I suggest that the people who do Linux Advocacy/marketing don't stop simply because everything didn't go perfectly. These numbers are only a defeat by microsoft if we let them be.
Remember, there is more than one part to the Linux community, just as there is more than one part to the body. Just as I wouldn't suggest that those who do documentation stop doing documentation and start coding, I also wouldn't suggest that those who do marketing stop and start coding.
It takes all sorts to run this world, and we shouldn't start neglecting any part of it, marketing included. Microsoft is going to try to put as much spin on these tests as they can. If we stand by and do nothing, we'll be as guilty as they are of the spinning. Sins of omission are still sins.
I don't suggest that we do it defensively, though. I suggest that we do it confidently and aggressively. act like we're in control. That's the neat thing about self-confidence. If you act like you know what you're talking about, people tend to believe you. So why should we give up acting like we know what we're talking about.
We've got some test results which prove that Linux can handle the needs of 99% of the world. Why exactly should we hush that up in favor of letting it be thought that Linux is slow.
It's so nice of Microsoft to pay for this apache advertising. Just as a point of reference, 1800 hits/sec is the same as 155,520,000 hits/day. I think that it's safe to say that noone in the world gets more than 150 million hits per day of static content. Wait, there's a better way:
1800 hits/sec * average 2k/hit * 8192 kbits/kbyte = 29,491,200 bits/sec, or 29.5 MBits/sec. What's that now, a T3 line? I know that a T1 line is 1.5 MBits/sec. Ok, so apache on one of these boxes can fille the equivalent of 19.6 T1 lines by itself. If (a bit more realistically, how many 2k files get those types of hits) those are just 10k files (let's not get into pictures), that's 147.5 MBits/sec, more than filling a T3 line, IIRC, and definitely filling up aapr. 98.3 T1 lines.
What's the problem with Linux/Apache, now?
May I suggest, if you can afford this sort of bandwidth, that you buy one of those 32CPU sun E10000 servers and call it a day? (or a server farm of linux boxes, since you're serving up static files.)
Oh, if you're serving up >1800 files per second of 2k files, who are you?
Oh, one more thing. If this is all on an intranet, you'll still need Gigabit ethernet if you're serving up the 10k+ files, so the sun box still applies to you.
That limit only exists on 32bit machines. 64bit Linux platforms, such as Linux/Alpha, have a truly astronomic file size limit, IIANVMM.
And who really cares all that much about the size of static binaries? If you're compiling static binaries, size obviously isn't a big concern.
Do even believe what you're saying? Do you really mean that Linux should be able to beat NT at utterly contrived situations that have little to no bearing on real life? Can you point me to one sight where people are requiring thousands of pages per second of static content for only six minutes?
I don't particularly care about samba/NT fileshare performance. That whole network neighborhood is so damn slow on the client end that a slow server wouldn't really be noticed. That thing is a pain to deal with when there are ten compuers on the network. I couldn't imagine how slow it would be with 100. Or the thousands that this test is trying to simulate. Anyhow, I don't really know much about file sharing, nor do I care much about file server performance. Other people can worry about that.
What part of this mindcraft test has any bearing on reality? It's a benchmark contest, pure and simple. It's an NT machine tuned for this benchmark versus what was a Linux machine tuned for bad performance on this benchmark. Now maybe it will be an NT machine tuned for this benchmark versus a Linux machine tuned for this benchmark.
To quote penguin computing's FAQ, "Now, this is a complicated question, and one that benchmarks can't really answer (unless of course, your intended application of the computer is to run benchmarks, in which case benchmarks will be a great indication of the performance you will get).
To quote someone else, "Disraeli was pretty close: actually, there are Lies, Damn lies, Statistics, Benchmarks, and Delivery dates." (found in fortune.)
Benchmarks are virtually useless, especially highly tuned benchmarks. What's the point in saying that if two OS's are configured in such a way as to achieve optimal speed for a useless application one can do that useless operation faster than another.
Static pages server to as many people as are being simulated using hardware with beta drivers has nothing to do with the real world. I defy you to come up with one example that has the sort of usuage pattern that was tested in the mindcraft test (i.e. >2000 page views per second of static pages for only 6 minutes).
So in a sense you are right, you can't dictate the external conditions for a server environment when making a server, one builds the server to those conditions. That's exactly what didn't happen here, the conditions were build for the NT server.
I congratulate the windows people that they can make an OS which can do so well on pointless benchmarks against crippled products. That's really something to be prowd of. Now pardon me while I go challenge my grandmother to a game of basketball...
It's good to know that in a corporate environment one only serves static web pages for 6 minutes at a time. And that one picks hardware that one knows is only supported by beta drivers. And that one doesn't tell the people who are trying to help you to optimize your system what's really on it, or, heaven forbid, let them actually see the box. That's what really happens in a corporate environment.
And no corporate environment ever moves over to NT on the client side, nor do they use win98 clients. They immediately reformat their hard drives and install win95 on all those new boxes that they get. Thanks for that wonderful description of what happens in the "corporate environment".
Don't ask me why I fed the troll. I really don't know. I just can't let that ever-so-slightly credible argument go unanswered.
How about a 6 hour test? That would certainly disable the ability of programs to hide performance hits over short periods of time. Does anyone really care how many hits their webserver can take in 6 minutes and then stop serving?
I remember hearing something about NT being configured to do all logging after the test, not during, while apache was configured to log during the test (with host names -> DNS lookup).
Not to mention, the chances of NT failing under a really heavy load in 6 hours is much greater than in 6 minutes, really helping things out to be more accurate. Noone runs their webserver for 6 minutes.
Remember that glide is being written by Daryll Struass (sp?) in his spare time without being paid. The voodoo people are basically using Daryll. Remember that in this case the "customer" is being given the drivers by Daryll, he's got no reason to besides the goodness of his heart.
If you want someone to whom you are the customer to write drivers, ask the makers of the voodoo chipsets to write their own drivers and do it on a timely basis for a funded company. Daryll's doing an amazing job considering that he's not being paid for it.
Please note that app->video card only works in the few instances when the app is simply writing pixel data (such as image viewers, etc.) and doesn't need any functionality from the card aside from being able to display passed pixels.
Most programs like things such as acceleration from video cards, at which point no sane programmer wants app->video card, that's just painful as hell as then you have to start supporting the acceleration of all different video cards, which is what X is around for in the first place.
So while what you're talking about is useful for unnacelerated video playback, it doesn't have much applicability to anything else. Wouldn't be that bad an idea, though, except that it would be a bit hard to work out the security model for.
It's called the MIT shared memory extension. Most programs that do intensive graphics stuff (TV players, the gimp, etc.) already use it. X and the program share a section of memory. The program writes to it, and then notifies X when it's done. Then X simply uses that area of memory for moving things to the card. The kernel isn't really involved at all except for setting up the shared memory area and for the pipe protocol to notify the X server. This also allows the X server to properly do clipping, and things like that.
I know that C2 doesn't mean much, but could you publish publicly this info that you have a bunch of standard slackware Linux boxes that have a C2 rating? It would be nice publicity for Linux, especially for those who have no idea what any of C2 security means.
The guy addressed this question earlier. They've got an old version of SunOS/Solaris which isn't Y2K compliant. They have to upgrade, and the Solaris licenses are quite expensive.
That's what someone said earlier, I disclaim any knowledge of knowing whether or not it's true.
On a different note, Sun boxes come with a real paucity of software, as far as I can tell. O fcourse, most of my experience is with older sun boxes, not newer ones, so I don't know if they do include all the utilities that I'm used to with Linux (such as compiler for more than 5 languages, make, bash, perl, etc.)
Note: I'm not advocating blindly replacing everything with Linux. It'll take a while before Linux is ready to be put on 64 CPU sun boxes (though I hear that there are people working on High end SMP stuff).
Actually, how water does tend to freeze faster than cold water, at least in some places. Hot water allows certain gasses to escape more quickly than cold water (I think that chlorine is one of the main ones) and thus can freeze at a higher temperature. At least that's what my earth science teacher tought me back in the eigth grade. I don't know if it's true or not, but something about certain saturation diagrams I saw in chemistry seem to point to that conclusion.
Just a side note.
Oh, btw, Xterminals were and are a good idea, but they're virtually unavailable nowadays. Besides, can you get any that have great acceleration for 1024x768 at 24 or 32 bpp and have 10/100 support for full duplex ethernet?
NFS is viable on trusted networks when you are root on the main computer. We use it in our computer lab, and it works quite well. On the other hand, I don't want to have to set up nfs exports to edit a computers /etc/hosts file, or something like that. And remote display works especially well for debugging X apps that work on your computer but not someone else's. As well, when you just need a quick and dirty solution to get something done without having the time to set up NFS and then take it down again, remote display is quite useful.
As well, remote display is good for thin clients, so that you can have a good graphics card and a cheap CPU.
On other big benefit, is that you can run other people's binaries. For example, I'm posting this comment from my Linux/Alpha box, in a netscape which is remotely displaying from my comp. sci. server (Dual PPro 200 Linux box). Quite convenient, really.
So yes, NFS can be advantageous in some circumstances, but remote display can be quite advantageous in others. It largely depends on the circumstances and what you want to do.
I think that he was just joking. There may be some special significance, though, tha tisn't terribly obvious.
There is currently a driver under development for the Matrox G200 cards. It's pre-alpha right now, but development is active and progressing (I'm hoping to contribute once my G200 PCI card shows up in the mail).
Do you have plans to support glx modules such as the current (development) glx module for the matrox G200 cards under Linux, and when it comes out (july-ish, I think) Precision Insight's glx DRI? It strikes me as strange that Q3 will require a 3DFX card since it's written in openGL? Btw, as 3D accel. for Linux gets better, such as the G200 driver, will you release AXP (Linux/Alpha) binaries, since we'll have 3D hardware acceleration as well? Thanks.
You might want to check out Cygnus's GNUPro utilities, which have lots of PII/PIII optimizations, I think. THey're fully GPL'd, too (not free, though, they cost $).
Do you run your NT boxes on NT certified hardware? I've seen rooms full of NT boxes, and they were semi-constantly being "rebuilt". In many people's experience, NT (and other Microsoft software) has a habit of comiting suicide. It has way too much of a reputation for being suicidal to ignore that.
I don't think that it's as much NT's kernel as NT's structure - being a black box (with hazy error messages), there's not much that you can do as a user to fix things. Linux (and UNIX in general) is very open, so when some library isn't working, you fix/replace the library, not the whole OS. When an NT library isn't working properly, your usual recourse is reboot/reinstall.
I'm not saying that there aren't some exceptions, but NT, especially to one is isn't a guru, isn't nearly as fixable as a UNIX box is (especially a Linux box).
Anyhow, this isn't meant as a mindless NT bashing session, but just poiting out that NT does have some problems in this area. About 50% of the NT boxes in a bank that I worked for had a habit of comiting suicide on a semi-regular basis. Note: this isn't as much those NT boxes running only on-CD stuff, but those running third party apps (like citrix, watermark, and securID). NT does seem to be fairly stable if you haven't installed anything that comes doesn't come on the NT CD.
The primary domain controller was really funny. It would continually keep eating up memory until it exhaused memory, then it would die. And it was a dual PPro 200 with 256 MB RAM. Pretty funny to watch, actually.
It's not the average Linux advocate, it's the average loud mouthed slashdot ranter who wants everything for free. Plenty of people, myself included, have paid for RedHat, and plan to continue to do so. They make a decent dist, and they give a lot back to the community.
It is nice that RedHat has their dist up for free, btw, for things like the comp. sci. department of my school where we're strapped for money and there's a semi-lengthy process involved in actually paying for anything, so for the computers in our lab we just download the RPMS and install.
But I (for personal use and at work), and many other people pay for RedHat. We're happy to. RedHat deserves the money.
Please don't confuse the loudest people with the most people. There's a world of difference, as the fact that RedHat is still in business and doing well proves. If it was only the minority of people who didn't want everything for free, how do you think that RedHat would still be financially alive?
1. In their scowering of the net, did they send mail to Ask Slashdot? I don't recall seeing an Ask Slashdot about configuring Apache optimally.
/tmp/ps.txt; sleep 10; done`. It looks like they either started swapping or as some other posters indicated ran out of file descriptors.
2. They should have posted their httpd.conf file, not given a few random tidbits from it. It looks suspiciously like they had apache configured as an inetd service, not running normally. The only way to tell, of course, is for them to post that file (and the other config file).
3. Why didn't they do the tests with the default parameters? Since they claimed that they didn't know how to configure apache, they should have tried the out-of-the-box config to see if they just screwed up. The Apache performance tuning page says that tampering with MinSpareServers and MaxSpareServers is unnecessary. They set them to 1 and 290, respectively. And they set maxclients to 290. I would like to see some actual records, like the output of:
`while true; do ps axm >>
4. Did they turn on apache's cache? They seemed to turn on the IIS cache.
What I'd really like to see is the results if they had just used a standard apache config, perhaps upping the MaxClients only, since they're planning on using more than 150 simultaneous connections, but they should have tried it without even messing with that.
Also, did they restart the apache server *after* the reconfigured it? They didn't document their actual test process. They also claimed to have tried different apache configurations. What were they? What was the performance on them?
Maybe Linux/Apache needs improvement, and maybe not. But this test contains way too little information to be fit for anything more than the recommendation that "more testing is needed". Did they even run more than one test for their results? Did they run a whole bunch of tests and just report the one that they want?
This test doesn't include nearly enough information for proper evaluation, and it isn't all that meaningful. I don't doubt their conclusion, that on the particular machines that they used with the particular configurations that they used they found that certain products on NT were faster than certain other products on Linux.
The real question is how representative were their tests? Can anyone really represent the information that they gave as being enough to tell? If not, that test doesn't merit anything more than a "that's interesting".
Does anyone know how transferable Linux knowledge is to UNIX? Of course, it depends on the UNIX that one's talking about, I know. I assume that Linux knowledge will transfer to a sys V based system better than to a BSD based system, since most Linux dists are based on sys V standards (correct me if I'm wrong).
Essentially, how well does being a Linux sysadmin prepare you for being a solaris sysadmin, or an AIX sysadmin, etc?
> As for the slavery, it is completely irrelevant,
> and there are so many things about that
> paragraph that are irrelevant that I am not
> going to bother to respond.
Interesting cop out. My question was pointed to determine your ideas of social morality, etc. There are is a small group of individuals would would say that the slavery should be legal, after all, the guy took out the morgage knowing what would happen if he got laid off. It's his fault for not developing enough skills to cope in a difficult world, etc. From what you've said, you seem to be one of them.
I want to know if you're one of those people who basically ascribes to the doctrine that the world doesn't owe anyone a living, so you shouldn't take care of anyone else. Note: I'm not attacking you personally. You're probably a kind, generous person. But the views that you expressed sound like the views of the dog-eat-dog proponent, that is, one who believes that because a dog can eat another dog, and sometimes will, that's a perfectly fine situation.
Most of us believe that some work should be done to make the world a better place. Sure, some people will steal if you let them, so you do your best not to let them.
You see, the problem with microsoft is not that they don't have the "right" to do what they've done. In the Hobbsian sense they have the right to do anything that they can get away with. On the other hand, they don't have the right to do what they've done in the sense of deserving it, or it being morally permissible. Just about all of the incompatabilities between M$ products and the rest of the world are artificial. They're designed into the system to act like a drug - to hook you and make you dependent on them.
Which makes me think of a side note - do you think that narcotics should be regulated? Do you think that crack dealers are simply "good businesmen" (they are selling a product for which there is a large demand)?
Anyhow, the argument against microsoft is that they have no right to compete artificially. Just like it should be illegal for microsoft to compete by filing hundreds of lawsuits against its competitors, or sending them hundreds of pounds of mail every day, or flooding their email box with messages, or ping flooding their competitors websites into oblivion, etc. It should be illegal for them to build addiction devices into their products.
But the problem with us coming to an understanding seems to be that your view of morality is "anything not directly violent is ok". Which works, in a sense. Of course, I can't see how, using that premise, you could support things like the civil rights movement, etc. Why should the government dictate that blacks have to be able to use the same bathrooms in private establishments as whites. Or why should blacks be able to sit in the front of the bus like whites in privately owned buses. Or why shouldn't a company be able to hire only white males solely because they're white males? It's their business if they want to do that.
I'm assuming that you're against equal employment laws, correct? And you believe that child labor should be acceptable because either the kid or his parents should be responsible for the kid. If he wants some extra pocket money, why should someone stop him. And I assume that loansharking should be perfectly legal. And that electricity and water companies should be able to charge whatever they want.
If the water company decided one day that they were going to charge everyone 100% of their assets for a drop of water, that's fine.
And there's nothing wrong with what OPEC did, is there? And I assume that price-fixing should be perfectly legal.
Out of curiosity, since you're dependent on your parents for food etc. in the first X years of your life, do they in effect own you? Should they be able to starve you, etc. if they want? Send you out naked into the snow if they don't want you in their house or wearing their clothes?
What do you propose to restrict? What laws do you say should be on the books?
There are lots of monopolies, many of which are strictly controlled. Look at your electric company (depending on where you live). Look at your water company. They're often monopolies, and often very highly regulated. It's not restricted to microsoft, nor was it invented by microsoft. The general principle is that if you can exert undue influence over your market, you're not going to be allowed to.
As far as Microsoft being required to publish specs, there is precident for that, as well. Ever go to the supermarket? Every food item which isn't composed of one thing only (i.e. bananas, apples, slabs of beef) are required, by law, to include both caloric information and the ingredients, listed in order of volume. Shouldn't companies be allowed to put anything they want into the food that they sell and not have to tell us?
Of course not. Why? Because then they place people in danger. If you don't know what you're dealing with, you might take something that you're allergic to. Why should M$ be forced to publish the full version of all their specs? Because without them, they can exert an undue influence on their market. If I write a story in Microsoft Word, I should be able to retrieve it, even if I no longer have Microsoft word. They have no rights to my intellectual property, and no right to keep it from me. By using an undocumented API, they are, in effect, keeping it from me.
Can you come up with a convincing argument for why selling something doesn't inherently imply selling the information on how to use it and interact with it? Can you give a convincing reason why a company should be allowed to make a product which hides my information, or the information that someone else gives to me, in a format that only they can read so that I am forced to pay them for it (without invoking the morality that if I steal from my neighbor but he can't force me to give it back, I should be allowed to keep what I've stolen)?
Think of it this way: do you think that it would be OK if Honda sells you a car, that can only use Honda ExpensoTires, and they're the sole manufacturer of them? And if you only found out about it afterwards (how many software boxes say "This product uses a proprietary format that binds you to our products and doesn't allow interoperability with other people, excepting that they also use this product")?
Out of curiosity, do you believe that slavery should be legal, if, say, people became slaves by selling their freedom rather than being cought or sold by someone else? E.g., Henry takes out a morgage. Henry is then laid off from work, and can't pay his morgage, so the bank reposeses his house. Unfortunately, property has devalued, and the house is only worth 80% of the morgage (and let's not forget interest). So the bank sells henry to pay for his debts. Do you think that this should be legal or illegal. If you think that this should be illegal, how do you reconsile this with your statement that microsoft should be able to do anything that they want, including product dumping, market control through proprietary APIs, etc.