Slashdot Mirror

Additionally I think it's good that IBM too have an interest in this area, since 1) competition is always good and 2) it makes for more accurate results. With some luck we can have peta-byte based grid by 2007.

I don't think that you are seeing the whole story. Basically, Intel has been holding out for IBM's silicon-on-insulator technology because it reduces power requirements a good deal. Unfortunately for Intel, IBM is pretty sneaky when it comes to licensing and often prefer to swap technology rather than accept cash. I'd imagine that IBM is holding out for an x86 cross-license agreement while Intel does not want to give that up.

What you've seen in the past couple years is a game of chess. With each move, the other hopes that they have positioned themselves to better reach a licensing deal. Intel's move to non-clock processor ratings was a big move in this game.

From what I've seen at Intel's developer forums, they're working on some radically different architecture. Something that isn't von Neumann at all. They're calling it "massively parallel" but the industry seems to think that this means multiple cores on one chip. I think that it means thousands or millions of "processing elements" on one chip (think really small processing elements). Their claim is that they'll be able to apply this architecture to everything from mobile to high-end servers simply by adding or subtracting elements as power constraints allow.

Where are my mod points when I really need them? I've lost count of the number of comments on this article that would get modded Overrated or Redundant if I had some right now.

Anyway, the lamers aside...compiling from source can have a number of advantages...the main one to my mind is stack protection.
I also like knowing that what is in the tarball I've downloaded is what is going into my binary...that there hasn't been a download/script kiddie backdoor add/compile to binary package procedure. It's like how I don't tend to go to McDonald's any more because I don't like the possibility of one of the people who work there spitting in my burger before it gets to me...same kind of thing.

I realise however that I'm trying to use rationality here to counter irrationality. The only reason why people here *really* don't like compiling from source is because it is clashes with the completely mindless, apt-getting "Debian IS Linux!" groupthink that I tend to observe on an almost daily basis around here.

I've often wondered why Slashdot's bias is so heavily tilted towards Debian in particular, actually. I'm assuming that for many of you it has a lot more to do with Debian having explicitly received Pope RMS's blessing than because of any genuine technical superiority. I find that deeply pathetic.

I think that once set up, Gentoo takes much less time to maintain and run than any Distro I have ever used (Redhat 5.2-9, Fedora 1 and 2, Debian, Mandrake, Knoppix (on hard drive), slackware).

I think that this is partly because the package manager is so great (trustworthy, stable and easy to use), and partly because the software that is installed is compiled specifically for all of the other software installed.

I agree - I've been through a similar list of distributions over the last 12 years and Gentoo is the first one that strikes the right balance of power and simplicity to me.

Two things got to me after a while with other distributions -- First, version upgrades of the distribution tended to be a complete pain. Moving from Mandrake 8.x to 9 basically meant backing up the old system, clearing off everything, installing Mandrake 9, and then selectively restoring the config of the backup. You *could* try it over the top if you really knew what you were doing, but you were going to end up fixing a lot of things after the install either way.

As others have mentioned here, there's really no "upgrade" to 2004.3 since you tend to keep everything up to date along the way with portage. I was in the middle of setting up a local rsync mirror and http-replicator for the Gentoo files tonight when the 2004.3 announcement came along, and the only thing I had to do was change my /etc/make.profile link because one of my boxes was based off the now unsupported 1.4 profile. And, of course, the instructions for doing this were displayed by the most recent emerge.

The second thing that bugged me about other distributions was that I always tended to need to install *some* package from source every now and then. Maybe because of a certain configuration on my system, maybe because the package wasn't available as an RPM for my version of Mandrake, etc. Once I started throwing tarballs into a Mandrake system, I lost the ability to handle the dependencies through rpm. I did create some SRPM's, but this was just too unwieldy at the time. I understand there is now a tool to capture a ./configure and make session to SRPM, but I don't know enough about it to know if it's all it's cracked up to be.

The first thing that really drew me to Gentoo was Daniel Robbins' article on IBM Developerworks about the ebuild system being designed to simply wrap around existing tarballs. While Gentoo was still a bit too bleeding edge at the time for me, I came back and tried it out when I got some time this year.

Sure enough, the ebuild system lived up to the writeup in the article from 4 years ago. When I finally found a package that I wanted that wasn't already in portage (well, it was, but it was far, far out of date), it took me less than 4 hours to learn, from scratch, how to write an ebuild file for it. The ebuild simply downloads the tgz from its normal website, untars it to portage's working directory, compiles it and its dependencies, installs it, and handles future package management for it. A very nice change from what I'd seen on previous distributions I'd used.

While Gentoo is certainly not the distribution I'd recommend to Linux newbies (mainly due to the lack of automatic install), portage is making package management and upgrades easy enough for me that I've installed it on four boxes so far in my home.

By "Power970" you seem to be referring to the PowerPC family. Don't forget that this chip family is based on the Power architecture from IBM (with some help from Apple and Motorola). The Power architecture contains other chips too, some of which don't have the limitations you cite. Certainly the chip architecture is fully capable of supporting machines with a larger number of CPUs running a single system image.

Although the really big (and custom) Blue Gene systems are apparently clusters, there isn't anything about the IBM Power Architecture itself that would prevent large monolithic systems from being designed and built.

The SPARC architecture can be used for machines like this, too. (Remember the CM-5?).

Building a supercomputer with a large number of CPUs running a single system image is a unique task with a limited client base, and SGI has experience with that. A whole lot more than CPU choice goes into making it work. The way they tell it it was quite a rush. The internal conversation must have gone something like this: "OK, team, we're going to build exactly one of these, and we already decided the price!" NASA doesn't build rockets like that, but SGI can build supercomputers like that. Impressive.

SGI deserves kudos. But if we step back and look at the big picture from the vantage point of SGI, it sure looks like SGI chose the IA-64 CPU for marketing reasons, not technical reasons. I'd have to guess that their engineering tasks would have been made easier by using a CPU that draws less power, for example. They've been on the ropes for years and conventional wisdom says to back Intel if you're in trouble because that's the safe bet for marketing. Why this remains conventional wisdom when the track record clearly shows that UNIX vendors who switch to Intel are cut up and fed to other UNIX vendors, is another topic.

You're right of course, that there are two different classes of super computers on the Top 500 list, with one class based on the cluster concept, and the other based on the concept of a single system image. Clusters are radically less expensive, and monoliths are better at certain computing tasks, and it's hard to compare them.

Monoliths often get custom case mods, though, and thus tend to look cooler. Who would hang a poster of a beowulf cluster of generic beige 1U rackmounts on their office wall? Everybody wants a poster of a Cray or a CM-5 or a Mach 5...

Hey! I just realized monoliths don't seem to look as cool as clusters lately. What's up with that?!

yes, specifically it looks like a system 360 model 65 , which first shipped November of 1965. The oldest machine I myself have worked on was a Cyber 175

Whitelisting is a simplifying concept, which one can understand more subtly as another factor to be accounted for in calculating probabilities, making your Baysian engine that much more efficient.

Someone with a valid stamp is less likely to be a spammer. Simply include it as a factor when calculating probabilities!

Or ignore the X-Hashcash field completely. As you choose.

If you read the article, you'd see that this was precisely the way in which SpamAssassin uses hashcash : as one factor amoungst many in a general system of spam classification.

Punchcards are making a comeback

Ahhh, consultants. Obviously they don't get Linux. And would never, ever recommend Linux on the Desktop.

Disclaimer: is my employer, but this is personal opinion.

Several thousand, actually, most of which are technical instructions on how to get [x] running on a particular Thinkpad model. Remarkably detailed, I've never had any trouble running linux on Thinkpads.

See also IBM Products Certified for use with Linux: http://www-307.ibm.com/pc/support/site.wss/MIGR-48 NT8D.html.

From yesterdays slashdot story it is coincidental that this article addresses the same issues that a lot of the problems the IBM scholarship award posed.

link to PDF of problems ,provided by MBCook yesterday.

You have to be inventive for a patent to stand up.

The best thing that AMD can have happen for them on the corporate front would be to get major vendors like Dell, HP, and IBM to offer their chips in their products.

IBM and Sun are already offering AMD based workstations, in addition to HP blades and supercomputers. At least at the workstation and server level, it seems as if the major vendors are already offering them.

Why are there no (or are there?) higher end CRTs with DVI inputs?

" IBM created its program in 2001 to drum up enthusiasm among students worldwide in Linux and open-source software."

It sounds like some of these "theoretical" challenges may be issues that they have in-house, and are looking for some free help to solve.

In 2001, post-bubble, I went on a job interview with a large, not-to-be-named corporate entity and was asked how I would approach / solve a few issues that they were having at the time. Wanting the job, I foolishly gave a couple insightful replies.

Did I get the job? No. Were the solutions implemented? A friend of mine who works in that group said that the hiring manager (he with whom I interviewed) suddenly had some great insight on their current problems, and proposed solutions that sounded quite a bit like my interview replies.

The bastard. My fault, though, for giving away the milk and not having them require that the cow at least be rented.

Is IBM doing the same with these kids? Dunno. Looking at the 2004 Official Rules, however, section 8 of the agreement reads:

By submitting entries, entrant grants the sponsors and their agents of this contest the right to publish, use, adapt, sell, edit and/or modify such entry in any way, in commerce, and in any and all media worldwide, including but not limited to the Internet, without limitation and without compensation to the entrants. Entrant also grants sponsors worldwide irrevocable, nonexclusive and royalty free right and license to use, have used, copy, reproduce, transfer, modify and/or display any information contained in their entry in
whole or part and in any form without compensation.

Things that make you go "tsk tsk." If I read this correctly, it means that they would own your idea as submitted, and can not only use it, but also sell it and / or patent it as their own.

It gets even more fun, in section 9:

At the sponsor's discretion, entrants may be required to submit source code to substantiate
results reported in the entry.

Bastards.

---

wwjd? jwrtfm!

Considering the last date for registration is Oct.31
What's the point?
See here's the lovely link that SHOULD have be hunted down by the editor: http://www.developer.ibm.com/university/students/c ontests/linux/
I mean I don't ask for much but if all you have is a spartan story on a contest you could at least type: "IBM Contest" into Google.
Sheesh, your job is to filter the news. That's it.

Second, I can't wait to see the results of this. Should be interesting to see how some of these are solved, and what other interesting challenges people come up with to try to solve.

IBM has been doing this for decades and they are exceptionally good at this. The difference is that, at least at this point in time, they do not actively do anything with their patents - at least not beyond the point of what's necessary to keep them. They just keep filing new patents to keep their asses covered. And, in a way, they have to do that to ensure the survival of the company. Think about it: it's way cheaper to just file for and receive a patent than to challenge somebody else's patent and to try and have that invalidated (something that hardly ever happens). It also helps with ligigation. If another company is suing you, you first check your database to see if they have violated one of your patents.

And to give you an idea of what I'm talking about, check out this quote from IBM's IP & Licensing website:

In 2003, IBM received 3,415 U.S. patents from the USPTO. This is the eleventh consecutive year that IBM has received more U.S. patents than any other company in the world. In addition to delivering these innovations through its products and services, IBM maintains an active patent and technology licensing program.

And, believe me, they're covering all their bases (last time, I checked they had 23k+ active patents and they have some exceptionally good lawyers). Please don't get me wrong. I'm not saying IBM is the bad guy here. I like the fact that they're supporting Linux as much as the next guy. I'm not even saying what they're doing is inherently evil. I'm merely trying to point out that patents are becoming a priority issue everywhere and that it's becoming increasingly important to CYA.

Ok, so of the three fastest computers in the world, one is almost exclusively dedicated to environmental climate models, and the other two have it as part of their tasks.

Perhaps this could bury the arguments on Slashdot that there is no hard data or serious research about global warming.

"...that used to be called Jalapeno that bootstrapped itself..."

Just to let people know, it's now called Jikes RVM and is still under active development. "RVM" is a Research Virtual Machine, which is like a standard VM, only researchers do weird, cutting-edge things to it (advanced garbage collection, advanced runtime optimisations). The idea is, one day, the standard JVM will pick up these changes.

Not related to the Jikes compiler.

There actually is an open source JVM from IBM, jikesrvm..

I was just reading an article on that. I understand that quantum teleportation works through a "Einstein-Podolsky-Rosen (EPR) correlation", rather than actual physical data moving anywhere, its a correlation between particle behaviors, right? Does that mean that its faster than light speed?

http://www.research.ibm.com/quantuminfo/teleportat ion/

http://physicsweb.org/articles/news/8/6/10

``doesnt ibm have a project named daisy that is a JIT vm running on top of the very machine it emulates?''

No. Daisy is a project that emulates other architectures (x86, PowerPC, and also JVM) on top of a VLIW processor.

Are there any similarities between this and what IBM does on their mainframes?

Regardless of the application, you need to have the keys. If you buy cards with the Visa test keys (usually only used for development and demos) then it would be pretty easy. I am not aware of any vendors that will arrange for secure unique key generation in small volumes, but I am sure that if you offered them enough money you could get some. If it is just for hobby applications then the keys might not matter and you could use the test keys. Still you need the tools to write, test, and load applets. You can get them from IBM here.

I really want to see SCOX got drained. They are pure evil. We must give some credits to IBM

• University Research and SCience, Rational Engineering, and Wisdom Entrepreneurial Database (URSCREWED)
• Dewey, Cheetham, and Howe
• The Mother of All Prior Art

maybe these guys. in fact, they sponsor events for precisely that reason.

maybe these guys. in fact, they sponsor events for precisely that reason.

Here is a summary of IBM's 15 year experiments with cosmic rays:

IBM's research on cosmic rays

I quote from this:

The cosmic ray intensity is greatest at high terrestrial altitudes, and approaches zero under extensive shielding. IBM has conducted extensive field testing3 of components at high altitudes (10,000 ft), at moderate altitudes (5000 ft), at sea level, and under shielding of 50 ft of concrete. All elevated-altitude tests showed cosmic-ray-induced fails in electronic components. In all tests, the observed fail rate scaled directly with the cosmic ray intensity, over a total observed change of more than 1000× .

There is also another related article at IBM.

IBM's research on cosmic ray densities at different places on earth

Osho

Here is a summary of IBM's 15 year experiments with cosmic rays:

IBM's research on cosmic rays

I quote from this:

The cosmic ray intensity is greatest at high terrestrial altitudes, and approaches zero under extensive shielding. IBM has conducted extensive field testing3 of components at high altitudes (10,000 ft), at moderate altitudes (5000 ft), at sea level, and under shielding of 50 ft of concrete. All elevated-altitude tests showed cosmic-ray-induced fails in electronic components. In all tests, the observed fail rate scaled directly with the cosmic ray intensity, over a total observed change of more than 1000× .

There is also another related article at IBM.

IBM's research on cosmic ray densities at different places on earth

Osho

Functionality is added to the card by securely loading JavaCard applets to the 1MB of storage. More info on JCOP can be found here.

everything gets thrown through SSH tunnels and I'd love to see an acceleration of whatever it is that SSH uses, as well as acceleration for creating those huge RSA/DSA keys we use all the time

Well, get an Opteron, install the tuned RC4 implementation and configure stunnel to prefer RC4 and you'll have no problems with throughput. The tuned RC4 won't speed up session startup because that's all public-key stuff. Large integer math libs could really benefit from tuning on 64-bit registers, though.

As far as key creation... the big problem is finding the large prime numbers and that's just plain slow. Even with a good source of random numbers, you still have to test each one to see if it's sufficiently prime. 64 bit registers would help some. I doubt you'll see general-purpose hardware that optimizes primality testing, though, since key creation is something you don't typically do a lot of.

I have a friend who worked (tangentially) with an Encipher box of some description... he said it was just a lump of aluminium with wires out the back and 3 blue LED's on the front, which later had another piece of aluminium welded over them since someone made a PoC that you could figure out the data being encrypted by the way the LED's blinked, or so the story goes.

It seems hard to believe a security device would be designed to show any information at all about the crypto operations, but I suppose it's possible. It is amazing how much information can be extracted by smart people who know a lot of math.

I'm more familiar with IBM's 4758 and related devices, which are very secure. If you ever want to see what extreme but very smart and methodical paranoia looks like, take a look at the 4758 design overview.

PowerPC is an architecture, its not just a proprietry brand like "G5" which everyone here seems fixated with. The G5 is just an *implementation* of the PPC architecture.

This is why PowerPC has supported 64bit, way before anyone but IBM actually *implemented* it.

Also, why would IBM make up a whole new ISA just for Sony, if anyone can license PPC?

PowerPC is an architecture, its not just a proprietry brand like "G5" which everyone here seems fixated with. The G5 is just an *implementation* of the PPC architecture.

This is why PowerPC has supported 64bit, way before anyone but IBM actually *implemented* it.

Also, why would IBM make up a whole new ISA just for Sony, if anyone can license PPC?

And I'm also certian that the US didn't just complete the first non-government manned space flight and doesn't have billions of dollars going to develop private space flight.

Give me a break.

China is emerging as an ecenomic powerhouse, and it looks like it will continue down that path, provided their government doesn't screw up. However please don't pretend like all good things come from China. I gave just a small list of the US companies that produce advanced hardware, including what drives almost all the devices you listed. Your MP3 player may be built in China but it's usually using TI DSPs and AD opamps.

You know it's perfectly possible for China AND the US to be economic powers, and for both to benefit from trade with each other.

AIX is not a pure micro based but something called an exokernel. I am not an expert on this but other slashdotters mentioned that exokernels are micro based but graphics are run in the kernel as well as a few other subtile differences.

I'm not sure those other Slasdotters are experts, either, as I've not seen "exokernel" used in the sense of "a small kernel to which user-mode servers are added to handle file systems, network protocols, and the like" - the sense in which I've usually seen "microkernel" used when it's describing something that's not a "modified microkernel" along the lines of NT's kernel or Darwin's xnu; it generally seems to refer to a kernel that provides very low-level primitives that provide a secure interface to the low-level hardware, with file systems, network protocols, and the likes implemented as libraries that use those low-level primitives. See the MIT Exokernel Operating System page, for example, or the Wikipedia article on "kernels" in the OS sense.

I've certainly not seen anything about an exokernel being like a microkernel except that graphics are run in the kernel; that sounds more like a "modified microkernel" to me.

But really it is a microkernel like design

Really? To which version of AIX are they referring? The current one for RS/6000s (and post-Great Renaming POWER series IntelliStations and p5 and pSeries servers), in its Kernel Extensions and Device Support Programming Concepts manual, speaks of a fairly traditional VFS interface (suggesting that file systems run in kernel mode, not in userland), speaks of "network kernel services" that at least suggest that network protocols run in kernel mode, and speaks of subsystems for various driver types suggesting that those drivers run in kernel mode as well.

IBM's had several different OSes named "AIX" that were not, as far as I know, the same OS. One of them - the one for the RT PC, I think - did, if I remember correctly what I'd read, have a "hypervisor" atop which the OS kernel ran, but I don't think the kernel in question ran in user mode atop the hypervisor, and that's not the flavor of AIX running on RS/6000's and POWER series IntelliStations/p5 and pSeries servers.

We're about to enter a really exciting time in computing. And you know what the Chinese said about that.

Thankfully when you need a new board, a link will come up to Epson.com to buy more.

In all seriousness, I find it interesting that this process cuts down "a large volume of photoresist, developer, etchants, stripping agents and other chemicals" needed for the process.

However, is this based on earlier processes or IBM's improvements in recent years?

In 2001 . "Michael Cummings, James Fuller, Jr., Timothy Krush, Mike Longo, Thomas Lyons, Curt Miller, Paul Speranza, William Wike, James Wilson, and Michael Wozniak of Endicott, New York, share $50,000 for developing and qualifying a new process that eliminated solvent use from the manufacture of ultra fine pitch (UFP) wire bond chip carrier products. A first in the industry, the team's innovations included the investigation and qualification of a dry film resist that achieves UFP's stringent photolithography specifications, while being compatible with existing printed wire board manufacturing steps. Benefits include, on an annual basis, avoiding 5.2 million pounds of chemical use, 5.6 million pounds of industrial and hazardous waste generation and off-site transport, and 110,000 pounds of process air emissions while saving over $5.6 million."

More info on here as well.

This is one of the things that has impressed me most about KDE. The protocol handlers can make working with some of these protocols a piece of cake.

Also worth noting however, is the DCOP system integrated into KDE. The protocol handlers and DCOP can and do make a powerful combination.

As with everything else in life, it may be cheaper to pay taxes than live life without them.

So this isn't a server then?

How about IPMI?

The marketing blurb goes something like this:

[IPMI] will allow for remote monitoring, management and recovery capabilities, regardless of the status or health of the server. New features such as enhanced security using leading authentication and encryption mechanisms in combination with remote console viewing will help reduce operational risk by securing remote operations. Moreover, with IPMI being implemented at the silicon level, it deals with monitoring basic server parts such as power supplies, fans, voltage and temperature irrespective of the type or health of the CPU or operating system.

Supermicro have a sub $60 daughter card (for their motherboards only) that seems to offer console access over LAN using 'out of band' bandwidth, whatever that is. There are other vendors offering 'IPMI-enabled' mobos as well.

Has anyone used one of these? I'm considering getting a few cards for the SuperMicros I colocate. And wonder how the seial console access works over a WAN, getting it setup, securing it, etc. And what support, server-side, there is for IPMI based monitoring.

--
Hosting VPS servers with SSH console access

Last updated 28-July-1999

When will people learn that MySQL, while a useful toy, just can't compare to a real Enterprise-grade database running on real Enterprise-grade hardware?

There is a tutorial for moving your home on that matter. I used it before but I find that tedious. Imagine if you have 14 users on a workstation and you want to migrate all of them. A little automation could migrate all users in a few seconds. I think Ubuntu would be the distro to introduce such a feature.

Here are some other interesting stories

http://www-106.ibm.com/developerworks/library/pa-m icrodesign/?ca=dgr-lnxw01fkjfdsa
http://www-106.ibm.com/developerworks/library/pa-m icrodesign/?ca=dgr-lnxw01dhifads
http://www-106.ibm.com/developerworks/library/pa-m icrodesign/?ca=dgr-lnxw01jj5324543
http://www-106.ibm.com/developerworks/library/pa-m icrodesign/?ca=dgr-lnxw01hdshdgh
http://www-106.ibm.com/developerworks/library/pa-m icrodesign/?ca=dgr-lnxw01jesusheadfuck