The guy is not the German minister of the interior, he is the Bavarian minister of the interior, and he is well known for this stupid publicity stunts. He is always at the forefront of demanding new databases, new surveillance, longer jail time for everything, more police and using the military to do police work. In short: the guy is a nutcase.
He is doing it to make sure people don't notice all the scandals his administration is involved in, for example they just completely botched a police IT spending bill, wasted millions on new software which does not work. And his law and order state had issues with soccer hooligans.
In short: the guy is a joke. Don't take him seriously.
I'm not going to invest in technology where it's not clear that my investment is safe.
And I am an early adopter, I just sunk 2000 Euros in a brand new graphics card, a 1920x1200 monitor, and a dual core CPU. So I can watch HD content. OK, and play some HD games:-)
They expect me to pay $1000 for a player that they can shut down remotely? They have GOT to be kidding me. No way in hell.
Oh, and they are targeting 40-50 Euros for a blank medium.
No. Thanks, but no thanks. Go try to butt fuck someone else, content mafia. I'll wait till you go bankrupt and then we'll see reasonably priced HD content.
I mean, they can't be THAT stupid, can they? What kind of value proposition is that? You can probably get HD content for $10 per month from the warez doods, without DRM, and without downscaling. This business model will be their downfall. And the people who say that if they fail, there will be no HD content, are wrong: every major cable TV provider has to offer HD channels these days. All the content will be there. So not even the pity argument counts.
Actually, under a "national security protection" program by the Pentagon, us black hat hackers traded two years worth of prison against half a year of Vista code review.
How would you teach security if not by trying out the attack tools?
I don't see what the hoopla is about here. He asked them to do a scan, not open them up and format the hard disk or download files on it.
Maybe his next assignment is the ethics. Maybe it's just a test to see if any of his students find this ethically wrong and refuse to do it. Maybe he would have given them extra points.
I run several servers on the Internet, and I get port scanned all the time. Even more so at home, where my dynamic DSL IP is hit by worms many times each day.
Dear American proto-hackers, you are welcome to come to Europe and learn the tools of your trade here. We meet every year between Christmas and New Year at the CCC Congress, and we have a LAN there, so people can get acquainted with the tools.
To call this issue well-known would be an understatement. It is even mentioned in the RFC2406 in the Introduction. RFC2406 happens to be to RFC that defines ESP mode in IPsec.
Had you bothered to read the blog entry yourself before commenting, you would have noticed that Kaffe and Classpath members are part of this project.
This does appear to be a consolidation project. We have several contenders for Open Source JVMs under Linux, but most of them lack in some way or the other compared to the Sun and IBM JVMs. So having one up-to-date one instead of five not-quite-there-yet ones is a step forward.
It's ridiculous how the Slashdot crowd is falling victim to Pavlov again.
If someone publishes a benchmark about your software, and finds out your software does not perform well, don't whine, don't behave like a child, don't start kicking and screaming, don't tear his hair out. Behave professionally.
Good starting points:
Does their test setup matter?
Can their number possibly be true?
What weak spots about the competition does their test reveal?
What can we do to improve the results?
Let me summarize what I think about their test. First of all, I believe their numbers. Apache sucks performance-wise, in particular if you run a busy site with dynamic content. That's why people are using squid in local accelerator mode before Apache. This is a good indication that some performance tuning is in order. But no, people rather wait for Microsoft to find out and then they start thinking about fixing it.
If this test was meant to be unfair FUD, they would not have tested TUX, just Apache.
But now to my questions above:
Question 1: is their setup relevant?
No. Sites who answer more than 5000 requests per second are not using a single web server, they are using a load balancer and a cluster.
Question 2: Can their numbers possibly be true?
The point I find least believable is that IIS had better CGI performance than Apache. Creating a process is really slow on Windows. Their result should be independently verified.
Question 3: What weak spots about the competition does their test reveal?
They did not test a single-CPU webserver (which is what almost everyone is using).
They did not test FastCGI or APAPI dynamic web pages.
So if we wanted to do a more balanced review, we would look at these.
Question 4: What can we do to improve the results.
Document APAPI better, I'd say. Almost nobody is writing their dynamic web page modules with APAPI. Everyone is using PHP or mod_perl. Benchmark Apache in real-world scenarios. Document best practices.
I hope they at least refrained from using berzerkeleydb as backend. I know several projects who have lost their repositories with bdb.
I personally have also lost data to bdb, but not in subversion. I would never be so mad to trust my source code to some broken database backend. How are you going to get your code back if there is corruption? strings?
The GNOME people are probably breaking out the champagne at this point.:-(
lame uses assembler code for vectorization. One of the new features of gcc 4 is the beginnings of a vectorization model. A good test for gcc 4 would have been to compile some C-only bignum libraries, and Ogg Vorbis! povray is also a good example, but then you need to test more than one specific test-run. Maybe gcc 4 makes radiosity in pov-ray 400% faster at a 2% cost in the rest of the code?
This guy is the Tom's Hardware of Linux reviews, except he doesn't have the annoying ads, and he does not split his lack of content over 30 HTML pages.
The new warnings of gcc 4 have helped me find a bug in my code. That saved me a week. Consider how much faster gcc 4 needs to make pov-ray or lame to save you a week of work!
gcc 4 can now reorder functions according to profile feedback. That should make large C++ projects faster. Also, the ELF visibility should make KDE start much faster. This should have been tested!
Please note that I'm not saying gcc 4 produces faster code. I don't rightly know. I do know it produces smaller code for my project dietlibc, where size matters more than speed.
The Inquirer is the greatest IT web site in the world, no question about that. But they rarely if ever do any reviews themselves. They just tell you trends and like to reviews done by other sites.
It is so rare that I find a review on a hardware web site that does not stink of bribery and/or incompetence, in fact, that I wrote down the URL for the one case that I noticed: http://www.sudhian.com/showdocs.cfm?aid=556
I salute journalists who value integrity higher than bribery. I'm not sure I could do that, if I were in their place. You have to feed the family, right? At least, here in Germany, we have c't, which is one of the last bastions of integrity in PC print journalism, and even they have their down sides.
The explanation for focusing on performance instead of battery power may also be that nobody cares about battery time. I recently shopped for a notebook and it was nearly impossible to get any real numbers on battery lifetime, even more so if you want it in writing.
In the end I bought an Acer notebook. The vendor told me it would hold out about 60-90 minutes on battery and would probably not run Linux. I am now using it on Linux and have 4 to 5 hours of battery lifetime.
As long as people only care for performance and nobody cares about battery lifetime, you can't blame AMD for this benchmark focus.
The fact that there is no real benchmark to get a number for battery lifetime is probably important to remember here as well. Playing a DVD in full-screen is a good test, for example, but who knows if you could gain more battery lifetime if you moved more of the decoding in the CPU instead of the graphics chip? In the end it's also a driver issue. And what do I care how long battery life is under Windows!
No. The definition of "open" in IT is: "open as in ``open your wallet".
And the source code to Solaris is open. You can get it and look at it, and fix bugs in it, if you like. It does not conform to the Debian pipe dreams, but heck, what does? Now that Debian considers removing drivers from their kernel which contain binary-only firmware images...
And "open standard" is a euphemism for a protectionist practice where companies get together and try to protect their market against outsiders. These standards usually codify the sum of the proprietary extensions of the people in the committee, and each of them has an obfuscated "reference implementation" and/or a few patents covering their share. That's why you will almost never find an open standard where some company is in the committee that does not have any leverage in the things required by the standard.
What we need is a real open standard, where only stuff is permitted in that is not covered by patents, can be freely used by everyone without any licensing claim by anyone, and where there exists a free software reference implemention. Then we can talk. Until then, all this "standards" talk is just trying to confuse you into believing you actually have a choice.
"Math intensive" means floating point intensive, because that is all the math normal people do with their machines. Calculating Pi to a billion digits is not floating point math, it is integer math.
The "math intensive" benchmark in this setup was Povray, and there the Athlon 64 shined. A lot. lame is also a floating point heavy application, and both CPUs are close there.
gzip measure memory performance. Apparently, the dictionary fit completely into the cache of the Xeon. Not a fair test.
I cannot comment on MySQL performance. It should measure integer and memory performance, I would wildly guess.
Bernstein's prime sieve is also integer arithmetic . If you have a prime with 100 million digits, the action is mostly in the CPU caches. Again, no fair test.
The unfairness of the benchmark setup becomes particularly obvious when you look at the chess benchmark. Chess (and other game AI type problems) do a lot of unpredictable jumps. That's the weak side of Pentium 4, and that's why Athlon 64 has historically outperformed Pentium 4s by a WIDE margin. Look at the hardware used by the PC chess tournaments and the chess grandmasters and you see Athlon and Athlon 64 all the time. If Anand now measures that Athlon 64 is outperformed by a Xeon, then the test setup can not have been fair.
I don't know about ubench, never heard of it before.
Password cracking and encryption is 100% integer arithmetic. And it is one of the mainstays of Opterons from the beginning. Anands measurement flies in the face of that.
He is the archetypical hacker, achieving the near impossible with shoestring budget and garage band manpower against all odds.
He is one of the quiet heroes of these days. Companies like Intel, IBM, HP have near unlimited budgets at hand, can afford to waste billions on crappy underachievers or stillbirths like Itanic, and still achieve but a fraction of what Centaur does.
The only gripe I have with them is that they don't put their CPU optimization guide plain up front on their web site so the gcc people can make a better viac3 target. So far, the proper gcc options are -march=i586 -mcpu=i486. And Via should put their data sheets, driver sources and specs on the web just like that, not behind some registration crap. Open source people are remarkably helpful if you don't put stones in their way.
I have never understood what those people are thinking when they publish.md5 files. I mean, really! If someone gets far enough to upload a compromised tarball, what stops him from also uploading a matching md5 file?
Exactly. Nothing.
That's why people with more than one brain cell upload.sign files. Those are digital signatures made with the GNU privacy guard. Digital signatures make sure that the guy who owns the secret key (and only him) can create signatures, which then everyone can check.
Of course there are also caveats (some dark three-letter agency could have cracked the key with their Roswell quantum computers, or someone could have stolen the secret key), but those are far less likely than some asshat uploading a md5 sum. Everyone can create matching md5 files for any content, but only I can create sign files matching my secret key.
So please someone hit those GNOME idiots with a clue stick, those md5 files must go. Now.
Oh, and while you are at it, please also tell the gnome people to use a directory structure where mirror programs (and people!) can see whether there were new uploads without having to recurse through the monstrous moloch directory tree from hell. Thanks.
Slashdot Mirror
Why are these people still in business?
I can't remember a single insightful thing they ever had to say.
Their predictions are usually blindingly obvious or wrong.
The guy is not the German minister of the interior, he is the Bavarian minister of the interior, and he is well known for this stupid publicity stunts. He is always at the forefront of demanding new databases, new surveillance, longer jail time for everything, more police and using the military to do police work. In short: the guy is a nutcase.
He is doing it to make sure people don't notice all the scandals his administration is involved in, for example they just completely botched a police IT spending bill, wasted millions on new software which does not work. And his law and order state had issues with soccer hooligans.
In short: the guy is a joke. Don't take him seriously.
How could they possibly reach 7.5 gbit with 6 Xserves, if each Xserve only has a 1 gbit ethernet connection?
:-)
This looks to me like some desperate attempt to justify the money they wasted on bad (Apple!) hardware, drugs and hookers
I'm not going to invest in technology where it's not clear that my investment is safe.
:-)
And I am an early adopter, I just sunk 2000 Euros in a brand new graphics card, a 1920x1200 monitor, and a dual core CPU. So I can watch HD content. OK, and play some HD games
They expect me to pay $1000 for a player that they can shut down remotely? They have GOT to be kidding me. No way in hell.
Oh, and they are targeting 40-50 Euros for a blank medium.
No. Thanks, but no thanks. Go try to butt fuck someone else, content mafia. I'll wait till you go bankrupt and then we'll see reasonably priced HD content.
I mean, they can't be THAT stupid, can they? What kind of value proposition is that? You can probably get HD content for $10 per month from the warez doods, without DRM, and without downscaling. This business model will be their downfall. And the people who say that if they fail, there will be no HD content, are wrong: every major cable TV provider has to offer HD channels these days. All the content will be there. So not even the pity argument counts.
So, Symantec, let's see the vulnerabilities you claim to have found.
Oh, you have none? It was just fearmongering to scare people into buying your products? I'm shocked, I tell you. Shocked!
This would be half as funny if Symantec products didn't open more holes than they close.
It's been a long time since he actually architected anything software-related, right?
Actually, under a "national security protection" program by the Pentagon, us black hat hackers traded two years worth of prison against half a year of Vista code review.
In retrospect it was not so good a deal.
How would you teach security if not by trying out the attack tools?
I don't see what the hoopla is about here. He asked them to do a scan, not open them up and format the hard disk or download files on it.
Maybe his next assignment is the ethics. Maybe it's just a test to see if any of his students find this ethically wrong and refuse to do it. Maybe he would have given them extra points.
I run several servers on the Internet, and I get port scanned all the time. Even more so at home, where my dynamic DSL IP is hit by worms many times each day.
Dear American proto-hackers, you are welcome to come to Europe and learn the tools of your trade here. We meet every year between Christmas and New Year at the CCC Congress, and we have a LAN there, so people can get acquainted with the tools.
Big noise, nothing behind it.
To call this issue well-known would be an understatement. It is even mentioned in the RFC2406 in the Introduction. RFC2406 happens to be to RFC that defines ESP mode in IPsec.
It must be a slow news day in Great Britain.
Had you bothered to read the blog entry yourself before commenting, you would have noticed that Kaffe and Classpath members are part of this project.
This does appear to be a consolidation project. We have several contenders for Open Source JVMs under Linux, but most of them lack in some way or the other compared to the Sun and IBM JVMs. So having one up-to-date one instead of five not-quite-there-yet ones is a step forward.
If someone publishes a benchmark about your software, and finds out your software does not perform well, don't whine, don't behave like a child, don't start kicking and screaming, don't tear his hair out. Behave professionally.
Good starting points:
Let me summarize what I think about their test. First of all, I believe their numbers. Apache sucks performance-wise, in particular if you run a busy site with dynamic content. That's why people are using squid in local accelerator mode before Apache. This is a good indication that some performance tuning is in order. But no, people rather wait for Microsoft to find out and then they start thinking about fixing it.
If this test was meant to be unfair FUD, they would not have tested TUX, just Apache.
But now to my questions above:
Question 1: is their setup relevant?
No. Sites who answer more than 5000 requests per second are not using a single web server, they are using a load balancer and a cluster.
Question 2: Can their numbers possibly be true?
The point I find least believable is that IIS had better CGI performance than Apache. Creating a process is really slow on Windows. Their result should be independently verified.
Question 3: What weak spots about the competition does their test reveal?
They did not test a single-CPU webserver (which is what almost everyone is using).
They did not test FastCGI or APAPI dynamic web pages.
So if we wanted to do a more balanced review, we would look at these.
Question 4: What can we do to improve the results.
Document APAPI better, I'd say. Almost nobody is writing their dynamic web page modules with APAPI.
Everyone is using PHP or mod_perl. Benchmark Apache in real-world scenarios. Document best practices.
I hope they at least refrained from using berzerkeleydb as backend. I know several projects who have lost their repositories with bdb.
:-(
I personally have also lost data to bdb, but not in subversion. I would never be so mad to trust my source code to some broken database backend. How are you going to get your code back if there is corruption? strings?
The GNOME people are probably breaking out the champagne at this point.
Uh, and this review is helping us... how?
lame uses assembler code for vectorization. One of the new features of gcc 4 is the beginnings of a vectorization model. A good test for gcc 4 would have been to compile some C-only bignum libraries, and Ogg Vorbis! povray is also a good example, but then you need to test more than one specific test-run. Maybe gcc 4 makes radiosity in pov-ray 400% faster at a 2% cost in the rest of the code?
This guy is the Tom's Hardware of Linux reviews, except he doesn't have the annoying ads, and he does not split his lack of content over 30 HTML pages.
The new warnings of gcc 4 have helped me find a bug in my code. That saved me a week. Consider how much faster gcc 4 needs to make pov-ray or lame to save you a week of work!
gcc 4 can now reorder functions according to profile feedback. That should make large C++ projects faster. Also, the ELF visibility should make KDE start much faster. This should have been tested!
Please note that I'm not saying gcc 4 produces faster code. I don't rightly know. I do know it produces smaller code for my project dietlibc, where size matters more than speed.
I'm not disputing their result, I am disputing their methodology.
A survey will be answered by suits in companies that large, and they tend to know nothing of the Linux installations in their company.
This survey would mean something if they asked the techies, the admins, the people who actually do something.
Also, many companies use Linux on WLAN APs or NAS solutions or the like without even knowing it.
I have professional dealins with many a mid-size company, and every single one of them has had some network service running under Linux somewhere.
It might be true that the management doesn't know, though.
The Inquirer is the greatest IT web site in the world, no question about that. But they rarely if ever do any reviews themselves. They just tell you trends and like to reviews done by other sites.
It is so rare that I find a review on a hardware web site that does not stink of bribery and/or incompetence, in fact, that I wrote down the URL for the one case that I noticed: http://www.sudhian.com/showdocs.cfm?aid=556
I salute journalists who value integrity higher than bribery. I'm not sure I could do that, if I were in their place. You have to feed the family, right? At least, here in Germany, we have c't, which is one of the last bastions of integrity in PC print journalism, and even they have their down sides.
The explanation for focusing on performance instead of battery power may also be that nobody cares about battery time. I recently shopped for a notebook and it was nearly impossible to get any real numbers on battery lifetime, even more so if you want it in writing.
In the end I bought an Acer notebook. The vendor told me it would hold out about 60-90 minutes on battery and would probably not run Linux. I am now using it on Linux and have 4 to 5 hours of battery lifetime.
As long as people only care for performance and nobody cares about battery lifetime, you can't blame AMD for this benchmark focus.
The fact that there is no real benchmark to get a number for battery lifetime is probably important to remember here as well. Playing a DVD in full-screen is a good test, for example, but who knows if you could gain more battery lifetime if you moved more of the decoding in the CPU instead of the graphics chip? In the end it's also a driver issue. And what do I care how long battery life is under Windows!
Actually, they are not drug related, they are drug prohibition related.
If all those drugs were legal, there would be no reason to shoot anyone about them, right?
uh, a liberal democracy?
What have you been smoking lately?
You might want to look at this,
this and you might want to read up on how they shut Al Jazeera down.
Actually, Japan has one of the largest military expenditures of all countries, topped only by the USA IIRC.
And even notorious agressors like the USA like to call their army "peace keeping force".
The difference is that Japan has actually refrained from attacking anyone in the last, uh, 30 years.
No. The definition of "open" in IT is: "open as in ``open your wallet".
And the source code to Solaris is open. You can get it and look at it, and fix bugs in it, if you like. It does not conform to the Debian pipe dreams, but heck, what does? Now that Debian considers removing drivers from their kernel which contain binary-only firmware images...
And "open standard" is a euphemism for a protectionist practice where companies get together and try to protect their market against outsiders. These standards usually codify the sum of the proprietary extensions of the people in the committee, and each of them has an obfuscated "reference implementation" and/or a few patents covering their share. That's why you will almost never find an open standard where some company is in the committee that does not have any leverage in the things required by the standard.
What we need is a real open standard, where only stuff is permitted in that is not covered by patents, can be freely used by everyone without any licensing claim by anyone, and where there exists a free software reference implemention. Then we can talk. Until then, all this "standards" talk is just trying to confuse you into believing you actually have a choice.
"Math intensive" means floating point intensive, because that is all the math normal people do with their machines. Calculating Pi to a billion digits is not floating point math, it is integer math.
The "math intensive" benchmark in this setup was Povray, and there the Athlon 64 shined. A lot. lame is also a floating point heavy application, and both CPUs are close there.
gzip measure memory performance. Apparently, the dictionary fit completely into the cache of the Xeon. Not a fair test.
I cannot comment on MySQL performance. It should measure integer and memory performance, I would wildly guess.
Bernstein's prime sieve is also integer arithmetic . If you have a prime with 100 million digits, the action is mostly in the CPU caches. Again, no fair test.
The unfairness of the benchmark setup becomes particularly obvious when you look at the chess benchmark. Chess (and other game AI type problems) do a lot of unpredictable jumps. That's the weak side of Pentium 4, and that's why Athlon 64 has historically outperformed Pentium 4s by a WIDE margin. Look at the hardware used by the PC chess tournaments and the chess grandmasters and you see Athlon and Athlon 64 all the time. If Anand now measures that Athlon 64 is outperformed by a Xeon, then the test setup can not have been fair.
I don't know about ubench, never heard of it before.
Password cracking and encryption is 100% integer arithmetic. And it is one of the mainstays of Opterons from the beginning. Anands measurement flies in the face of that.
I call bullshit.
He is the archetypical hacker, achieving the near impossible with shoestring budget and garage band manpower against all odds.
He is one of the quiet heroes of these days. Companies like Intel, IBM, HP have near unlimited budgets at hand, can afford to waste billions on crappy underachievers or stillbirths like Itanic, and still achieve but a fraction of what Centaur does.
The only gripe I have with them is that they don't put their CPU optimization guide plain up front on their web site so the gcc people can make a better viac3 target. So far, the proper gcc options are -march=i586 -mcpu=i486. And Via should put their data sheets, driver sources and specs on the web just like that, not behind some registration crap. Open source people are remarkably helpful if you don't put stones in their way.
Go, Centaur! Go, Via!
What an excellent point to make!
And how non-obvious!
I mean, who would have thought that whether blue actually looks blue might be a teeny weeny little bit important as well?
Sheesh. And I thought Slashdot was inane before.
I have never understood what those people are thinking when they publish .md5 files. I mean, really! If someone gets far enough to upload a compromised tarball, what stops him from also uploading a matching md5 file?
.sign files. Those are digital signatures made with the GNU privacy guard. Digital signatures make sure that the guy who owns the secret key (and only him) can create signatures, which then everyone can check.
Exactly. Nothing.
That's why people with more than one brain cell upload
Of course there are also caveats (some dark three-letter agency could have cracked the key with their Roswell quantum computers, or someone could have stolen the secret key), but those are far less likely than some asshat uploading a md5 sum. Everyone can create matching md5 files for any content, but only I can create sign files matching my secret key.
So please someone hit those GNOME idiots with a clue stick, those md5 files must go. Now.
Oh, and while you are at it, please also tell the gnome people to use a directory structure where mirror programs (and people!) can see whether there were new uploads without having to recurse through the monstrous moloch directory tree from hell. Thanks.
Sheesh. Now that wasn't so hard, was it?