Slashdot Mirror

Security Systems: http://www.zoneminder.com/
http://www.linuxjournal.com/article/8513
http://linas.org/linux/secure.html

Alarm Systems: http://www.voip-info.org/wiki/view/How+to+implement+an+alarm+system+with+Asterisk+and+a+webcam
http://www.linux-support.com/cms/diy-burglar-alarm-system/
http://www.linuxjournal.com/content/interfacing-disparate-systems

CCTV: http://www.tuxradar.com/content/build-your-own-surveillance-zoneminder
http://www.seattlesurveillance.com/

Building Automation: http://www.sciencedirect.com/science/article/pii/S092658050500097X
http://dl.acm.org/citation.cfm?id=1029022

Engineering Apps: http://loll.sourceforge.net/linux/links/Software_Applications/Science_-_Engineering/index.html

You get the idea I hope. So what can't run on Linux?

there are several GNU accounting apps available for linux. i've used a couple for tracking/auditing my bank accounts. http://linas.org/mirrors/www.aerospacesoftware.com/2003.06.21/GNU_Cash_for_Business_users_Howto_Guide.html is the first hit on google...

as for CAD - i used to be a structural engineer before i got into technology and yes, if you MUST use autodesk's autocad then you're screwed. but there are several GNU CAD products out there; if you are willing to learn them. and anything you need to open in autodesk can simply be converted. even if you have to buy ONE copy to do the conversions... however, the problem i ran into was there were no mathcad or RISA-type products available. but there are plenty of GNU FEA/FEM softwares and excel based solutions that *most* anything you needed to do could be done in linux.

so, its not impossible.... it just takes a lot of research and re-training you;re entire staff to be able to use linux+these other softwares. which is a fairly tall order....

Well, imo? Not really - how does "pervasive message passing" function as a COMPLETE OPPOSITE, in terms of design

It doesn't. I will give this one more try...

"Pervasive message passing" is a feature of language, or the framework. It's like object-orientedness, or garbage collection. It's a property of the tool.

"Coarse multithreading", as you've explained it, is a methodology. So is "fine-grained multithreading". These are like, oh, model-view-controller. You can implement model-view-controller with an object-oriented language, or, if you really want, you can implement it in a purely-functional language (like Haskell), or in a language without even a concept of a subroutine (BASIC with GOTOs), etc... you can do it in a language with garbage collection or without...

What's more, "coarse" and "fine-grained" are relative and a matter of opinion. How do you define a "task"? And keep in mind that while having thousands of threads sounds like "fine-grained", they are not sharing data, except by passing messages back and forth...

This is all & again: IF NO CHANGES ARE MADE? This NEVER happens, period.

Exactly.

Threads additionally (again), are much "lighter" to instance, than an entire process (as forking does, vs. spawning extra threads).

Sigh...

Apparently, you still don't get it, even though you just tried to explain copy-on-write to me. fork() on Linux is implemented as copy-on-write, and has very little overhead. Linux sees threads and processes as essentially the same thing.

The overhead of using processes instead of threads is pretty much irrelevant, especially if you were going to use message-passing in the threads, even moreso if you're intentionally doing "coarse multithreading". Of course, you could also use shared memory between processes.

And yes, all of this is Linux-specific. Forking a process on Windows is much slower.

A few - but, the point is NOT so much about them

No, the point is that, originally, I said that I would not see any real improvement if I had a quad-core system instead of dual-core.

it's more about what happens when you have a "CPU Hog" of an app...

You seem to be operating in a fantasy world where:

1. I have more than one app that wants to saturate a core
2. I do not have the ability to set the priority of these apps lower
3. I absolutely need my one app to run at 100% of that one core, rather than 90% of it. (Especially ludicrous as you keep bringing up Amarok -- if Amarok is using 100% CPU because of a bug, forcing it to use 90% isn't going to cause problems.)
4. I was saying that dual-core is actually useless to me -- it's not. It just means I don't have to do item #2, above.

I never said or implied that multicore would never be usable, or that it was useless to everyone. I was merely speculating that most people are probably in the same situation I am.

Furthermore, most of the points you bring up, I knew, but didn't think they were relevant to the discussion. Had you been paying attention, you might have noticed that I was specifically making a statement about how completely irrelevant the deep technical details are to the point of whether dual-core is useful to the general population -- most people I know would never notice a machine they were using was dual-core until you pointed it out to them.

It'll take some reading and combining from multiple sources. I've been doing it for a few years, combined with a handful of upgrades, plus setting it up as an iSCSI backend- all of that lent to the pool of greyness in my head.

I recommend Gentoo to do this with. Other distro's dont include the latest mdadmtools required to manage and migrate RAID5 md devices. Ubuntu is catching up, I believe.

Here are some places to start:

http://gentoo-wiki.com/HOWTO_Gentoo_Install_on_Sof tware_RAID
http://www.gentoo.org/doc/en/gentoo-x86+raid+lvm2- quickinstall.xml
http://linas.org/linux/Software-RAID/Software-RAID .html
http://linas.org/linux/raid.html
http://evms.sourceforge.net/
http://www.tldp.org/HOWTO/Software-RAID-HOWTO.html

It'll take some reading and combining from multiple sources. I've been doing it for a few years, combined with a handful of upgrades, plus setting it up as an iSCSI backend- all of that lent to the pool of greyness in my head.

I recommend Gentoo to do this with. Other distro's dont include the latest mdadmtools required to manage and migrate RAID5 md devices. Ubuntu is catching up, I believe.

Here are some places to start:

http://gentoo-wiki.com/HOWTO_Gentoo_Install_on_Sof tware_RAID
http://www.gentoo.org/doc/en/gentoo-x86+raid+lvm2- quickinstall.xml
http://linas.org/linux/Software-RAID/Software-RAID .html
http://linas.org/linux/raid.html
http://evms.sourceforge.net/
http://www.tldp.org/HOWTO/Software-RAID-HOWTO.html

150kW using photovoltics requires about 1000 sq metres of space in the middle of the desert at high noon. You'll need about 4000-5000 sq metres

Call Center, Bug Tracking and Project Management Tools for Linux

http://linas.org/linux/pm.html

I'm curious too. After all, Linux can certainly run on an IBM VM/ESA 390 and zSeries. However, perhaps the thought is that Linux is not as scalable as versions of UNIX, not that it isn't scalable at all. On the other hand, others disagree and think it's an ideal approach.

It's true. Lately I've been noticing that living in the open source universe can be a lot like attending high school: everything is a popularity contest. If you're not one of the "cool kids" you don't get any attention, even if what you're working on is more mature, more sophisticated, and just plain better than what they're working on.

What I'm about to say is probably not going to be taken well, but here goes anyway: Slashdot is probably the "football team / cheerleading squad" of the open source high school -- the place where the coolest of the cool get the most concentrated doses of glory and attention. There are certain people (whose names I shall not defame in this post, lest I get moderated down to -99 or something) who could make a stupid remark about how they think it would be better if people didn't wear matching shoes, and Slashdot would run half a dozen stories about it.

The best example of unsung heroes might be Linas Vepstas. He wasn't one of the "cool kids" so the world pretty much ignored his project, which was to port Linux to IBM mainframes -- he actually got it working, for the most part. IBM ignored his work and went it alone, and nobody knows much about Linas Vepstas now.

Unsung heroes indeed. Let's all try to avoid making open source a fashion show. Most of our best technology was built by nerds, and nerds aren't known for their social skills.

"The Internet reacts to censorship as damage and routes around it." - John Gilmore (frequently misattributed to Howard Rheingold)

Good luck! Cheers, Joel

Okay, how about this time we wait until AFTER they start using the algorithm before we tell them it's been hacked. I'm looking at you Edward Felton. ;-)

Star Wars Kid strikes again.

Raid is a backup solution as much as having a backup is a backup solution. You will lose all your backups sooner or later with roughly the same probability that you will lose the ability to recover your RAID.

If downtime is your biggest problem, use RAID-1 (plain-vanilla mirroring) so you can just switch to RAID-0 (non-RAID operation) when one fails.

And I take issue with this statement:
RAID is not a substitute for frequent, regularly scheduled backup.

It's predicated on the idea that the humans are in error, but human error will hose your backups, too, in ways you never dreamed ("no, I didn't notice the tape wasn't moving...it's always just made that clicking noise, and I've been here for years" (gestures to huge rack of tapes...)). If you didn't make a verified backup before installing the RAID in the first place, you're too stupid to live.

Swap, schmop.

Put them all in the box and get a RAID running on them and USE THE ERROR CORRECTION modes (R1 or R5 depending on whether you want to keep the most speed or space).

You'll never have to "back up" again, because your data is backed up automatically with every read or write.

RAID is also available for Windows.

Swap, schmop.

Put them all in the box and get a RAID running on them and USE THE ERROR CORRECTION modes (R1 or R5 depending on whether you want to keep the most speed or space).

You'll never have to "back up" again, because your data is backed up automatically with every read or write.

RAID is also available for Windows.

Linux supports software raid, too.

If you believe in Newton, sure free will doesn't exist. If you do not, then a good argument can be made for free will using various non-newtonian physics.

Not to mention, console, ssh and snmp work just fine. -- Curtman

Call Center, Bug Tracking and Project Management Tools for Linux.

Also: Problem Management Tools Summary.

Load up Movable Type and blog those volunteers into submission!

Also, this site may be helpful. Someone has already done a LOT of research and typing.

There are several other good databases around. InterBase looks pretty excellent.

I'd be interested in RainingData (<span face=poker>now there's a truly inspired name change</span>) nee Pick GPLing their MultiValue database. Overall, a pig to use and maintain compared with something like PostgreSQL, but it still does some neat tricks and has a reasonable community around it.

I know folks that use wwwreq and it seems to work as well as the people you have actually handling the tickets.

Also, see this page for a bigger list.

Exactly right.OFX is slated for inclusion into Gnucash.

There is one hitch, however: download method. Some financial institutions require your application (MS Money/Quicken) to download the information from your bank, while others provide a browser-based hyperlink download of the qif/ofx file for you to import into your application. As long as your app supports the file formats provided, no vendor/financial institution is needed.You can curently import qif files into Gnucash.

The setup in which you download the files with your fincance application requires your app to first connect to the vendor's "branding" server, which then redirects you to your financial institution to begin the download -- this is where vendor involvement with the financial institution gets involved.

I suppose that if someone knew the address to their FI's download servlet, their application could be written to go directly there and bypass the branding servers. However, I don't think that FI's usually publish the URI for their download servlets so getting this info could be a challenge (though tech support should be able to tell you).

I've found that this interactive download method is more error-prone and resource-intensive (for the FI) than the browser-based options.

Here's a comment on building an OFX parser found on the Gnucash project goals pags:

There are two ways to build an OFX parser. One way is to build a compile-time DTD parser that treats the DTD as if it were an IDL, and generates C language stubs for a parser. This approach was attempted and abandoned because it leads to fragile C code and a very large binary.

• The parser is fragile because minor DTD non-compliances are hard to parse, handle and recover from.
• The parser is huge because the DTD results in hundreds of (C++) objects being generated.

The other method would be to perform run-time DTD parsing. This is attractive particularly because it is a more commonly-used approach; there are a variety of XML tools available that provide this function.

There already is a S/390 emulator... now all you need is the OS... or you could be daring and try Linux on it.

One reason: Lotus Notes. Granted, there are also open source alternatives out there (such as for instance Tutos), but that doesn't help you much if you are an employee at a company which uses Notes. Wine allows you to run Linux on your workstation while still being able to access the corporate document and discussion databases.

Of course, it is in IBM's power to show their true commitment to Linux by making this point moot with a native Linux Notes client, but for some weird reason they don't want to, despite their Linux commitment in many other areas...

Thanks, finally someone who understands compared to others assume suggesting COM is a trick to sneak MS technology into Linux.

I'm not for destroying extensibility. Instead, I would love to see a system where objects are installed onto the server, maybe even at run-time off the web if an application needs it. Even better, a system where a developer can extend an existing control to add new functionality or build a new control off of generic objects.

Just a suggestion, if you guys don't like my ideas, I won't lose any sleep. Just trying to help.

Oh, and the driver thing too, don't forget that.

Ozwald

Anyway: GNUCash is *NOT* a business accounting system; the GNUCash people have said so themselves. It may become one in the future, but it is not right now.

Humm, first generation unix partitioning from IBM, or 5th generation partitioning from Sun

IBM has been doing LPARs and Virtual Machines a long, long time on mainframes. According to this article, IBM started studying logical partitioning in 1976. I think Sun qualifies as the newcomer if you are just talking about partitioning technology.

See Linux on the IBM ESA/390 Mainframe Architecture for info on running Linux on S/390 LPARs

I would suggest looking into implementing NPC instead of OFX support though. Just my opinion on this, but OFX sort of sucked for us to support. I won't go into technical (possibly proprietary reasons), but I believe NPC is more flexible. Don't quote me on it, just throwing out my worthless $.02 for general consumption.

On another note, since we do have sort of a merger thing going on with our old rival (now partner) Transpoint, which is a joint venture with Microsoft and First Data, it might be more advantageous to go with the OFX model for overall scalability, but that's more of a question for our sales departments I guess.

There's a very good reference on Call Center, Bug Tracking and Project Management Tools for Linux at http://linas.org/linux/pm.html. It has helpful comments, too.

There's a very good reference on Call Center, Bug Tracking and Project Management Tools for Linux at http://linas.org/linux/pm.html. It has helpful comments, too.

http://linas.org/linux/pm.html has some good general information regarding Call Center, Bug Tracking and Project Management Tools for Linux, including open source solutions.

I have heard a lot of good reviews about KeyStone, which is free to download for personal use (including source code!) at http://www.stonekeep.com/content/download. But StoneKeep isn't open source-enlightened yet, so they still charge $$ if you use it commercially.

Also, the S/390 code is only really useful with binary-only drivers for lcs (3172, old OSA) and qdio (GB OSA). Until these drivers are available in full source, i wouldn't call the port complete ;-)

--jochen

It really doesn't matter whether SDMI is hacked before or after music starts being released with either the Phase 1 or Phase 2 watermark. However, all the watermark can do is identify a track, and give some sort of guidelines of the rights the 'licensee' has. ** The whole plan hinges on software that pays attention to the watermark. ** SDMI's plan is simple: Divide and conquer. The 'big' software companies pay attention to the watermark because they are rich targets for lawsuits, plus they want to use popular (i.e. big five label) content. After the big software companies have signed up, SDMI can start going after the small players. Having a link to Gnutella on your site might make you a target! Let's get anonymous, reliable transfer and hosting for open source audio software going! Chris Owens San Carlos, CA

The Anti-Mac ideas are only part of a revolution that will be as mind-altering as the internet was. The seeds are already planted: David Gelertner described some of the concepts. ERights, Publius, FreeNet, Napster, the Eternity Service, distributed.net, seti@home, google.com, rufus.w3.org & etc. implement some tiny subsets of these ideas. But there's far more to come. See Eternity for details.

• software cost is minimal, compared to the
• real big money charged for tailoring

The benefit here is that the industry is already accustomed to paying for the follow up service. The license cost is the entry price, and if it's $0, then smaller organisations may come in. They know that it's not an off the shelf product that anyone can just install (sound familiar?).

I could imagine a group of people getting together, taking on the open source management of the ERP (BAAN or whatever, maybe even working on the current projects such as http://linas.org/linux/xacc/projects.html or like the business specific attempts such as http://linudent.sourceforge.net/ and http://freshmeat.net/appind ex/1999/10/05/939153658.html) and then giving it away for free and making money from the tweaking. Start your own SAP killer, targeted at small businesses and build up from there...

The trouble with the world is that the stupid are cocksure and the intelligent are full of doubt.

Take a look at Call Center, Bug Tracking and Project Management Tools for Linux, which has a bunch of these.

Gnu Bug Tracking System (Gnat) has a billion (or so) user interfaces you can use with it - Tk, web, command-line, etc.

Even the high-end call-tracking systems like remedy can be (and often are) configured to email you when you get a new case assigned to you, but these are hugely expensive and it doesn't sound like you wanted to spend a lot of money.

RT, Keystone, and php Helpdesk would be good starting points.

Browsing through the comments, it seems that most of the posters don't have a clue.

There are many applications for Linux on a S/390, not one of the least being the ability to provide thousands of virtual machines on one physical machine. Just think what this could do for the dedicated hardware web hosting business!

With an IBM S/390 I could provide all of the capabilities of the dedicated machine, including full root access to each user of a virtual machine, and I could do it for under $100 per month. I'd make a killing. I'm suprised someone hasn't already done it.

• Bigfoot -- Linux on a S/390

• Previous Slashdot articles:

• IBM Ports Linux to S/390
• S/390 Support now in Kernel 2.2
• The practical value of Mainframe Linux
• IBM runs 41,000 copies of Linux on Mainframe
• Linux on S/390 gaining momentem

A list of problem tracking software.

I hope this comes about soon. I've become quite reliant upon online transactions, etc. I can't make the switch till they get OFX in (I even made a half hearted offer of help once, but received no response).

But what are they going to do to develop linux?

The list is long. Very long. Do you know about Jikes, for starters? how about Linux on the 390? It goes on and on.

I honestly don't see how the UDI is going to help. I can't help but think that it's just commercial Unix vendors waking up to the fact that Linux actually supports more hardware than any of the commercial Unixes. When was the last time you sat down at your Linux box and said, "Darn, I really wish Linux had a driver for foo. SCO Unix has a driver for foo, I wish I could use that."

It simply doesn't happen.

Intel has come up with a clever hack that would allow Unix systems to share drivers, and now they want Linuxers to actually do the grunt work and write the drivers. It would be different if Intel had a whole bunch of UDI drivers for hardware that Linux didn't already support, but they don't. They are simply hoping to tap into Linux's talent pool.

As for the Linux port to the S/390, it appears that their are actually two of them. The reasons the non-IBM version was written can be found here. The same site says that the official IBM version was done "for political reasons."

What is a mainframe from the point of view of linux? What sort of changes are required to port to a mainframe, particularly to a S390? Folks talk about stuff like IO channels and VMs but what are the salient features that I don't know about?

I/O channels: at least on System/3x0, I/O is done by constructing a "channel program", which is a series of commands whose opcodes tell the channel, and the peripheral attached to it, to perform some operation (read data, write data, search a disk track for a block whose "key" has a certain value, rewind a tape, etc.) - there's also a branch instruction (Transfer in Channel) and, as I remember, some ability to do conditional skips over channel commands. The CPU just issues instructions such as "start I/O" to start a channel program; the channel program does the data transfer.

VMs: one "meta-OS" running on S/3x0 mainframes provides, to the OSes running atop it, a "virtual machine", that looks much like a real S/3x0, and whose disks might be subpartitions of the real machine's disks, whose communications controllers might be part or all of the real machine's communications controllers, whose system console might be the terminal on which the operator of that virtual machine is logged in, etc.. (VMware is somewhat like this.) Linux could run on one of those "virtual machines", and one of the other OSes for S/3x0 could run in another one, so that you can run applications for Linux and applications for, say, OS/390 on the same machine, without having to port a UNIX application from Linux to OS/390 (which has a UNIX environment - but it's not completely like the UNIX environment to which most UNIX programmers might be used, e.g. it doesn't use ASCII as its native character set, it uses IBM's EBCDIC).

Assuming I'm familiar with von Neuman architecures, stack machines, microprocessors, minicomputers, memory mapped memory, memory mapped devices, IO ports, interrupts, and the unix concepts of streams, char devices, block devices, etc... what don't I know about mainframes?

Well, IBM mainframes have a fairly conventional von Neumann-architecture instruction set (CISC, 16 32-bit general-purpose registers, variable-length instructions, memory-to-memory character/decimal arithmetic instructions, memory-to-register and register-to-register binary arithmetic instructions) with some specialized add-ons. The CPUs in them are, these days, microprocessors implementing that instruction set.

I don't know if OS/390 implements memory-mapped files, but the hardware certainly permits it - it has a fairly convention in-memory-page-table MMU.

I/O devices aren't memory-mapped, however; you tell them to do things by telling a channel to send them commands. The channel program can interrupt the CPU either to say that it's finished or, as I remember, to notify it that it's reached a certain point in the program.

The UNIX I/O model isn't what OS/3x0 has traditionally implemented, although the UNIX environment atop it implements that, and a Linux port would implement such an I/O model.

One thing I *do* know from using them briefly, is that IBM "terminals" (3270s? something like that) are really weird: they are not simply connected via a serial cable. They have these extra control signals that light up indicators that say "you can't type now, I'm busy" and the text editors seem to do their editing on the "screen" locally and then send the changes back when you are done. I realize this has nothing to do with the kernel, but it would seem to make the whole experience quite surreal.

How surreal was your experience posting your article? You presumably filled in the text in the "Comment" box, doing any editing locally, and then sent the changes to Slashdot's server when you were done by pressing the "Submit" button.

I remember, several years ago, reading some magazine in which somebody described much of the Web as "3270 for the '90's". A lot of the stuff with HTML forms and HTTP POST operations resembles the way I think 3270's work.

The instruction set and I/O architecture of S/390 is described by ESA/390 Principles of Operation. Links to that and some other manuals can be found on the Linux on the IBM ESA/390 Mainframe Architecture page.

Does anybody know where I can find the gcc patches to compile for S/390?

S/390 Linux, or S/390 OS/390? In either case, there are links from the Linux on the ESA/390 Mainframe Architecture page.

If you are looking for the kind of clustering that Windoze NT does, then you want something like TurboCluster Server from TurboLinux, it is clustering for high availability and high throughput for web servers. TurboLinux

If you need more general load balancing clustering for enterprise applications, look at Linas Vepstas's Linux Enterprise Computing pages at http://linas.org/linux/, he has a section on clustering on that page.

If you need supercomputer numbercrunching or render-farm type clustering, then the Beowulf approach is what you want. Linas' pages also have a section on Beowulf type clustering.

As far as revealing the inner workings, if someone really wants to know that they could read the OS/390 Principles of Operations on IBM's bookmanager site.

Or they could just follow this link if they don't want to go searching for it.

The Linux on the IBM ESA/390 Mainframe Architecture page has links to various documents about S/390.

I bet they: 1.) port the gnu compiler

It's apparently already been done. (I think S/370 support has been in there for ages; the link is to somebody offering pre-compiled binaries for OS/390 UNIX services.)

2) Offer Linux under VM as a standalone OS. VM runs other operating systems really well. Both of these would reduce the porting effort because the actual hardware would be hidden.

I was under the impression that the "virtual machine" that VM implemented looked rather S/3xx-ish, complete with virtual channels talking to virtual I/O devices that look somewhat like real S/3xx I/O devices.

• The 390 has a killer instruction set for doing I/O more efficiently (with fewer CPU cycles) than any other cpu out there. Sure AIX/irix/slolaris can do XXX TB/second i/o but what good is it if the cpu is 100% busy? The mainframe can do this I/O wich the cpu 99% idle.
• The 390 allows the disk drive cables to be hundreds of meters long and so you can fill a whole room with disk. You can't do this with an SP/2. You can try using NFS disk servers, or a SAN fabric running SCSI but you'l surrender performance, or number of peripherals.
• The Linux/390 page says it all: 65536 devices and all of them busy
• LPAR allows you to boot multiple operating systems. This allows you to create test regions and production regions on the same box, and acheive uptimes that are out of the ballpark.
• You don't have to reboot the OS to add more disk or even add more CPU's. Does your favorite unix box even allow you to plug in a CPU without turning off power?
• Address spaces: the instruction set allows 16+ address spaces that all have access to priveldged (kernel) instructions. This allows kernel deamons and kernel modules to each run in thie own address space, without corrupting other parts of the kernel. They could even do an OOPS without taking down the rest of the kernel with them. Its kind of like having "capabilities" designed into the CPU instruction set. There is no other unix/risc-box on the planet that allows you to install a kernel module without compromising the security of the kernel itself. This is exactly how some recent Linux security breaches ahve occured: some cracker got root shell and installed a secret back door as a kernel module.
• The multiple-address space ability can make client-server and corba really really blaze. No other CPU out there has this ability. All the other CPU's require you to use pipes or sockets or shared mem to do stuff like this. Imagine having a 100% secure syscall without the overhead of pipes/semaphores/mutex's/shmat's!

P.S. lets give credit to the website where this work is happeneing: Linux/390 Its nice that IBM is hyping this, but IBM is *not* pumping actual $$$ into this, the way that e.g. SGI is pumping $$$ into Linux. They're just taking all the credit :-|

• The 390 has a killer instruction set for doing I/O more efficiently (with fewer CPU cycles) than any other cpu out there. Sure AIX/irix/slolaris can do XXX TB/second i/o but what good is it if the cpu is 100% busy? The mainframe can do this I/O wich the cpu 99% idle.
• The 390 allows the disk drive cables to be hundreds of meters long and so you can fill a whole room with disk. You can't do this with an SP/2. You can try using NFS disk servers, or a SAN fabric running SCSI but you'l surrender performance, or number of peripherals.
• The Linux/390 page says it all: 65536 devices and all of them busy
• LPAR allows you to boot multiple operating systems. This allows you to create test regions and production regions on the same box, and acheive uptimes that are out of the ballpark.
• You don't have to reboot the OS to add more disk or even add more CPU's. Does your favorite unix box even allow you to plug in a CPU without turning off power?
• Address spaces: the instruction set allows 16+ address spaces that all have access to priveldged (kernel) instructions. This allows kernel deamons and kernel modules to each run in thie own address space, without corrupting other parts of the kernel. They could even do an OOPS without taking down the rest of the kernel with them. Its kind of like having "capabilities" designed into the CPU instruction set. There is no other unix/risc-box on the planet that allows you to install a kernel module without compromising the security of the kernel itself. This is exactly how some recent Linux security breaches ahve occured: some cracker got root shell and installed a secret back door as a kernel module.
• The multiple-address space ability can make client-server and corba really really blaze. No other CPU out there has this ability. All the other CPU's require you to use pipes or sockets or shared mem to do stuff like this. Imagine having a 100% secure syscall without the overhead of pipes/semaphores/mutex's/shmat's!

P.S. lets give credit to the website where this work is happeneing: Linux/390 Its nice that IBM is hyping this, but IBM is *not* pumping actual $$$ into this, the way that e.g. SGI is pumping $$$ into Linux. They're just taking all the credit :-|

This exact same story was posted by Roblimo last week on Sat Nov 20. In that very discussion, I posted a comment detailing how this was Old News, and that Linux Today ran a related story the month before that!

For those who want all the links in one comment: The Linux Today article referenced an article in the Danish version of ComputerWorld, and the comments on LinuxToday pointed out this project's home page.

I knew something was funny when the story link for this article was black instead of green like it usually is. Can you moderate an article as redundant?