Slashdot Mirror

Of course IBM was - for the exception of some token R&D projects in Austin, they sent all of their R&D and IT jobs overseas. The only things left here in the States are salesmen and management and other assorted business support services.

IBM has become just an Indian tech reseller.

Then explain these job opening totals:

Engineering (hardware): USA 67, India 0, China 17, Other 47
IT & Telecommunications (non consulting): USA 233, India 181, China 113, Other 574
Research: USA 125, India 4, China 8, Other 29
Software Development: USA 126, India 50, China 320, Other 468
All categories combined: USA 2781, India 615, China 1043, Other 3596

Or are you claiming the R&D is outsourced to other companies (as opposed to working for IBM overseas), in which case I'd have to ask which companies? As far as I can tell IBM owns more pure research facilities than most companies. Note that three out of the eight are in the US, while no other country has more than one.

there will always be a system administrator with the technical ability to snoop data stored or in transfer. The only reason you can slam Google here is because they actually caught the guy.

Not always, if security is properly implemented. Google implemented poor controls, and had no idea he was doing this. There can be preventative controls and monitoring and auditing, they implemented neither, they only figured this out when the parents complained.

Their security is poor, simple as that. Now granted, many companies probably have lousy controls also, and the hackers (internal and external) seem to be always a step ahead, _however_ there are companies that are doing this right, Google is not one of them.

Big machines, not toys.

Yours In Moscow,
Kilgore T.

If you direct to the IBM announcement, which mentions the system in more detail then this linked article - http://www-03.ibm.com/press/us/en/pressrelease/32414.wss The New zEnterprise 196 " From a performance standpoint, the zEnterprise System is the most powerful commercial IBM system ever. The core server in the zEnterprise System -- called zEnterprise 196 -- contains 96 of the world's fastest, most powerful microprocessors, capable of executing more than 50 billion instructions per second. That's roughly 17,000 times more instructions than the Model 91, the high-end of IBM's popular System/360 family, could execute in 1970." 17k x improvement in performance in 40 years? I suppose that is about right...

Yeah, I suppose Wikipedia is full of dreck. But their time-line seems pretty accurate. Other than your general disdain, do you have anything else to offer?

These dates seem to match fairly well
http://books.google.com/books?id=fRvbxgH4wmsC&pg=PA7#v=onepage&q&f=false

From IBM is a pretty similar analysis of the economics and the significant role of the iMac: http://www.ibm.com/developerworks/power/library/pa-spec7.html

Enter the iMac.

The original "bondi blue" iMac was the first computer to offer USB ports without offering "legacy" ports. That's right -- no serial ports, no ADB. This changes the network effects. Before the iMac showed up, there were many millions of PC users who had no USB ports and perhaps a couple of million who had a USB port and also legacy ports. The biggest market in 1998 was in serial and parallel ports (or joystick ports, PS/2 ports, and so on) -- there was no reason to target the USB market. That would just restrict your audience.

The iMac presented a ready-made market of users who chose the Mac line for its graphics capability. In turn, the iMac offered a captive audience of users who would buy a USB peripheral but would not buy any other kind of peripheral. These users provided a market for USB peripherals that wasn't facing competition from other port choices. The result was a flood of USB devices in white-and-blue plastic. This was a crucial turning point that created a reason (tied to a proven system choice) to prefer USB to non-USB ports.

Once adoption was foist onto this substantial segment of users, the technical merits of the technology won out easily. USB's technical superiority (for most peripherals) to the conglomeration of a half-dozen different port types was unambiguous.

Well, I'm not sure that this is a technology question: IBM has made a 200 ppi screen (*) a long time ago,
which was quite expensive as expected as it was a new technology, but the curious thing is that no other
LCD maker tried to compete on resolution, so prices didn't go down!

I thought at first it was because Windows XP couldn't use correctly high resolution display,
but Windows Vista and 7 are supposed to be able to use them and there's still no vendors trying to sell
high resolution display for consumers.

*: http://researchweb.watson.ibm.com/roentgen/

Ok, first of all, how are you going to talk about 'startups' doing all the 'innovation' then go on and on about Apple, a company that's been around since 1977? Oh, wait, I forgot. Everything before OS X 10.0 was just a dark phantasmal nightmare of beige plastic and doesn't count.

Second of all, the likes of Apple don't create core routers capable of moving 322 terabits per second. They're also not creating electronic chess grand masters, are they? Nope. But at least they're shiny!

Disclaimer, I'm writing this on a MacBook Pro that I'm fairly fond of. It's a nice machine. It's hardly ground breaking or innovative. It has some nice features, and it looks pretty, but frankly I, think being able to move 322Tb/s through a router is a little more earth shattering than a fucking music player.

There is effectively only one Java implementation, the one controlled by Sun/Oracle. Sun killed most of the others early on with legal threats, and the few remaining ones seem to fail to meet the conditions of Sun's public patent grant. Anybody who writes Java software is pretty much stuck with running it on Sun/Oracle's proprietary implementation or its nominally GPL derivative. You're joined at the hip with Oracle, in the bending over kind of sense.

Um, no, not necessarily.

This doesn't mean you're wrong, but it means that you can't use these links as citations.

IBM is shipping a closed source implementation and they have licensed Java, so therefore their closed source implementation is not proof of the existence of an independent implementation not created under license. How much code it shares doesn't matter, what matters is that IBM has a license and is bound by the terms.

If you want to state that independent, conforming implementations exist, you need to show that some entity that is not a licensee has created a conforming implementation.

For IBM we know anyway: just download the software https://www.ibm.com/developerworks/java/jdk/, read the licensing information, and then look through the class files and implementation-dependent parts.

I had recently started poking around the lguest hypervisor. From my limited reading I believe 2 of the 3 memory subscription choices mentioned in the article are present in Linux. Existing linux based open source hypervisors like kvm etc use paging/swap mechanism (i,e, for x86 - the paravirt mechanism). Ballooning is possible using the virto_balloon. Kernel shared memory in linux allows dynamic sharing of memory pages between proceses - this probably doesn't apply to virtualization.

I couldn't find any CPU over-subscription thing in open-source hypervisors. It seems to be the only area where open-source hypervisors are lacking.

On an other note, established players like IBM tend to use Type-1 hypervisors (link) for enterprise servers, it would be interesting to see how this company fares against them.

Have a little config file consulted by the shim library:

~/.bwnice.conf: 199.237.54.1:80 5 MB/s /tmp/videocache.bin 5 MB/s etc...

I dunno what your use case is, you could use this for stuff you start as a user, or added to the system config, and have it apply to all users. To do system-wide stuff, tc is already plenty good enough, thought it only applies to network stuff. Dunno of anything for limiting b/w to local disk.

anyway, you don't need any kernel anything for this feature. basic rule is if it can be done in user space, it very often ought to be done there.

http://www.ibm.com/developerworks/linux/library/l-glibc.html

http://www.linuxjournal.com/article/7795

I'm generally fine with QWERTY on the desktop/laptop keyboards as I didn't find Dvorak to be significantly better, but QWERTY is completely suboptimal for input when you're using one finger/stylus or, say, two thumbs. I've played around with the FITALY keyboard on my WM PDA, and although there was learning process, it did seem to be a somewhat faster. Also the Metropolis layout, based on a Monte Carlo method of the same name, looked very promising, but I don't recall ever finding an implementation for WM, and I was too lazy to make one myself.

It's kind of sad that all of this seems to be fading now that everyone's trying to reanimate QWERTY for touchscreen use with fat fingers, since if some of the autocorrection methods (which I personally dislike) were applied to these optimized layouts, we could get some amazingly fast and accurate input methods.

Market

IBM mainframes dominate the mainframe market at well over 90% market share.[6] Unisys manufactures ClearPath mainframes, based on earlier Sperry and Burroughs product lines. In 2002, Hitachi co-developed the zSeries z800 with IBM to share expenses, but subsequently the two companies have not collaborated on new Hitachi models. Hewlett-Packard sells its unique NonStop systems, which it acquired with Tandem Computers and which some analysts classify as mainframes. Groupe Bull's DPS, Fujitsu (formerly Siemens) BS2000, and Fujitsu-ICL VME mainframes are still available in Europe. Fujitsu, Hitachi, and NEC (the "JCMs") still maintain nominal mainframe hardware businesses in their home Japanese market, although they have been slow to introduce new hardware models in recent years. The amount of vendor investment in mainframe development varies with marketshare. Unisys, HP, Groupe Bull, Fujitsu, Hitachi, and NEC now rely primarily on commodity Intel CPUs rather than custom processors in order to reduce their development expenses, and they have also cut back their mainframe software development. (However, Unisys still maintains its own unique CMOS processor design development for certain high-end ClearPath models but contracts chip manufacturing to IBM.) In stark contrast, IBM continues to pursue a different business strategy of mainframe investment and growth. IBM has its own large research and development organization designing new, homegrown CPUs, including mainframe processors such as 2008's 4.4 GHz quad-core z10 mainframe microprocessor. IBM is rapidly expanding its software business, including its mainframe software portfolio, to seek additional revenue and profits.[7][8]

Quit being a hypocrite and go after every mainframe manufacturer. And yes, I am calling you a hypocrite and a troll.
Nor does IBM force you to use a specific operating system. From the IBM web site of all places. http://www-03.ibm.com/systems/z/os/

Linux on System z Show descriptions | Hide descriptions Featured topic Increasing economics for server consolidation Linux and z/VM benefit from the enormous improvements of the IBM zEnterprise 196 server capabilities in the areas of consolidation, security, reliability and disaster recovery. You can "do even more with less". 10 Years Linux on IBM System z For the last decade, clients around the world have benefited from the strengths of Linux on System z. Learn more about Linux on System z

I don't see companies like Apple advertising the fact, and also offering to help install other operating systems on their hardware.

Show me in witting, on the IBM web site, where their software can not be run on mainframes built by other manufacturers. I don't want your blog, I want IBM official restrictions. Otherwise, your a troll. Nothing more, nothing less.

Dell should be OK for servers.

A few years ago in a prev company, I found that IBM's x336 servers were better designed and of better quality than the "equivalent" Dells. And they weren't really that much more expensive. The Dells weren't crap, they just weren't as good. Cheaper, less reliable and layout not so good. Noisier too.

But I think the Dell servers have improved in design - they look comparable now:

IBM: ftp://public.dhe.ibm.com/common/ssi/pm/rg/n/xso03094usen/XSO03094USEN.PDF
Dell: http://www.dell.com/downloads/global/products/pedge/en/server-poweredge-r610-tech-guidebook.pdf

When I wrote "SourceFire site", you read "snort.org" because...?

Are you going to keep showing how stupid you are, or you think it's time to stop?

Please go to http://www.ibm.com/ and try to find their opensource projects on the front page.

Dont be so sure about that... resently on slashdot it was linked to IBM's new project with a computer doing jeopardy.

here is the link again: http://www.research.ibm.com/deepqa/

It's even worse than that. You can go down to the local grocery store, buy a $5 Visa/MasterCard gift card, and register it online with ANY details you'd like. This is also why you should not rely on a credit card for age verification.

Not only that, but the last time I looked at credit card protocols, they didn't even have the capability of sending the cardholder's name! Street address? Yes. Zip code? Yes. CVV/CVC? Yes. Name? No.

Want proof? Check out IBM's VisaNet API. There's no way to pass in a customer name. Or you could look at the actual protocol (although it's served off archive.org so it's sllloooowwww and unreliable).

What I imagine might be going on is that some credit card processors provide an API for sending a name, but the name gets dropped once it hits the credit card network. The merchants THINK the name is being verified, but instead any old name is silently accepted.

They have the zIIP and zAPP processors on the z series mainframes, which are specialty procs. zIIP for database and encryption, zAPP is basically a java VM in hardware. IBM is big, and they have specialty fabs to make silicon for specialty mainframes. Yeah, they are expensive but worth it for some applications.

I was thinking the same thing. OLAP is all about manipulating big 2d and 3d sets, blending them with other sets, etc. All things GPUs have ops for on the die. Not that there aren't already relational db accelerator chips in the mainframe arena (such as the zIIP). Obviously the drivers and front end needs to be remade to make the programming make sense, like OpenDL (data language) instead of OpenGL.

eFUSE is a chip technology developed by IBM is a special type of chip where the code isn't completely static- based on the operation of the device, an eFUSE can blow itself. This can reroute the logic in a variety of ways, or be used as a self destruct mechanism.

It's reversible, but only by Motorola directly via JTAG. They have the custom code needed to flash the chip back to its original state.

Sun changed the patch access policy early 2006, years before Oracle bought Sun. http://www.mofeel.net/1208-comp-unix-solaris/8197.aspx Ever since then you needed to have a support contract to legally download Solaris patches.

No, you just had to have a sunsolve account. It did not have to have an account tied to it (see the second comment on the link you provided).

It also is not true that IBM provides patches for free - you cannot download AIX itself for free, duh. AIX also doesn't run on non-IBM hardware. So you have to pay IBM at least twice before you can even use AIX.

IBM indeed provides patches to their AIX operating system without even logging in. Check out this fix central link for AIX patches:

http://www-947.ibm.com/systems/support/supportsite.wss/supportresources?brandind=5000025&osind=5329283&taskind=2

You are correct about not being able to download AIX for free, however when you purchase hardware you get a free license to run AIX on it. And with free patches you are not required to maintain vendor support in order to maintain it.

Thank you - came to post this.

Here's a cut and paste of an email I sent about 8 months ago about this issue.

Some of the information has changed (about fedora and RHEL6), and obviously the fact that "no one manufactures a drive larger than 2TiB".

Note: I was pretty careful to use TB and TiB when I meant TB and TiB; and to use physical / logical when I meant physical or logical, and to use partition / filesystem when I meant filesystem. Words mean things, and understanding why "the 2TB problem" is a problem is not difficult but it involves reading and understanding the concepts.

Feel free to copy pasta this if you need it.

WHY IT DOES NOT WORK:

Master Boot Record limitation.

It's nothing to do with partition sizes or file systems or partition
types or ext3.

ANY system that uses an MBR (i.e. ALL RHEL, including Fedora 12, which
is what RH6 is supposedly based on) will NOT boot off of a logical disk
of 2TiB or more. I'm sure that goes for physical disks as well, but
currently, no one makes a 2TiB disk (the 2TB disks out there for
consumer use are slightly smaller than 2TiB).

The reason for this is that the MBR contains the logical block
addressing information for the logical disk on which it resides, and
this is stored in a 32 bit integer. ANY computer that uses a logical
disk that is >2TiB cannot use an MBR to boot* from that logical disk,
because the MBR cannot address the whole disk. (*without a crowbar)

So, before any information about the formatting or filesystem or any of
that stuff is loaded, you turn the computer on, the bios takes over, and
then seeks the MBR to go to the next boot step.

WHAT WE'VE TRIED:

Certain raid levels can create a logical disk of arbitrary size;
however, raid1 and raid10 cannot create logical disks except along
physical disk geometry. This was tried with PERC 6's. Raid5 will allow
for logical disk sizes along arbitrary boundaries, IIRC.

You may be able to use sfdisk to manually designate a partition table of
a size which would truncate your usable space below 2TiB of logical
disk, and sacrifice the end of the logical disk. But we couldn't get it
  to work.

WHAT WILL WORK:

If you're going to *boot* off of a disk that is >2TiB, you *MUST* use GPT.
There's an overview of GPT here:
http://www.ibm.com/developerworks/linux/library/l-gpt/index.html
Basically, it's an MBR replacement that is fully backwards compatible.
It does NOT require EFI bios, although it was developed to work in
conjunction with it.

GPT, however, means using GRUB2. Redhat 5.x does NOT support grub2 (and
thus, neither does current Anaconda). Redhat 6.x likely will not
support grub2 by default, although Fedora 12 has grub2 in its repository
with the standard **THIS IS NOT STABLE USE AT OWN RISK** disclaimers.

Alternatively, you can move the MBR to a smaller logical disk. It would
be easy to have another disk in the system that's of insignificant size,
and put /boot and the MBR onto that disk, and have the (rest of the) OS
on a logical disk that is multiple TiB. Finding a filesystem that
supports these giant sizes is less easy - ext3 could only do it with
bigger blocks - but systems such as reiser or XFS should be able to
handle it.

So, in order to do this, there are basically 3 options:

1.) Hack anaconda to use grub2 and parted instead of fdisk.
2.) move the MBR anywhere else - a 128mb USB key would work
3.) (*this is the crowbar) - hex edit the MBR to stop at the 32 bit
boundary and forego the space at the end of the drive above 1.999999 TiB.

Anyway, just wanted to make sure that everyone had the benefit of our
efforts. Most of you probably already knew this, but some might not have.

Actually, both of you are making distinctions that are meaningless. The true goal was to get bought out by IBM, same as Platform Solutions Inc

IBM Corp. and plug-compatible mainframe startup Platform Solutions Inc. (PSI) moved their battle from the courtroom to the negotiating table, and now Big Blue plans on buying its onetime adversary.

Since late 2006, the two have been engaged in a lawsuit in which IBM sued PSI for patent infringement on its z/OS operating system. In early 2007, PSI countersued, alleging that IBM had shut out competition by coupling z/OS with its own hardware.

Since then, motions have been filed back and forth, but nothing has been settled. Until Wednesday, July 2, that is, when IBM announced it would acquire privately owned PSI. Financial details of the deal have not been disclosed

Sound familiar?

The difference is that PSI had some proprietary stuff that IBM could use. Turbo Hercules doesn't so no buy-out.
from the USPO

United States Patent Application 20060085599
Kind Code A1
Woffinden; Gary A. ; et al. April 20, 2006
Processing of self-modifying code in multi-address-space and multi-processor systems

Abstract

A method and system of storing to an instruction stream with a multiprocessor or multiple-address-space system is disclosed. A central processing unit may cache instructions in a cache from a page of primary code stored in a memory storage unit. The central processing unit may execute cached instructions from the cache until a serialization operation is executed. The central processing unit may check in a message queue for a notification message indicating potential storing to the page. If the notification message is present in the message queue, cached instructions from the page are invalidated.

From IBM: http://www-03.ibm.com/press/us/en/pressrelease/24560.wss

ARMONK, NY - 02 Jul 2008: IBM (NYSE: IBM) today announced it has acquired Platform Solutions, Inc. (PSI), a privately held technology company headquartered in Sunnyvale, California. PSI's technologies and employees will become part of the IBM System z business unit of the IBM Systems and Technology Group. Financial terms were not disclosed.

PSI's technologies and skills, along with its intellectual capital, will become part of IBM's long-term mainframe product engineering cycles and part of IBM's future product plans.

"IBM's strategy is to continually evolve our mainframe technology to help our clients tackle the most demanding business issues," said Anne Altman, General Manager, IBM System z. "We will continue to move the mainframe forward through both IBM innovation and by acquiring new technologies. We welcome Platform Solutions, Inc. and look forward to collaborating with them."

"We are pleased to become part of IBM, knowing IBM has the industry's most comprehensive vision for the future direction of enterprise computing, and has the requisite technologies to realize that vision," said Michael Maulick, President and CEO, Platform Solutions, Inc. "This acquisition makes the most sense for our companies -- to collaborate on future technology offerings and maximize our combined knowledge and skills for the benefit of IBM clients globally."

As part of this acquisition, both IBM and PSI dropped their respective claims against each other.

Prove it.
How about years of investment designed to meet the needs of very high volume and very high reliability... Reservations systems, banking systems (the one's with your account balance), telephone orders and equipment systems.... these are all $1M/minute for down time Bet-the-business systems that never quit, never fail, and do millions of transactions per day.

The serious people that run these shops know how to do cost/benefit analysis... they still use IBM equipment because it best meets their needs. Fur instance, while at Qwest, deploying changes to five sets of main frame applications for DSL service, I leaned about their GeoMax product - the DWDM fiber optic channel to connect two data centers, each with SysPlex mainframe clusters, for microsecond sync should the business have an outage of a full data center - spending millions per year "extra" to avoid $Ms/minute outages.

http://www-03.ibm.com/systems/z/advantages/gdps/index.html
IBM Geographically Dispersed Parallel Sysplex (GDPS)

Yes, at my current bank client we are tied to COBOL code that would take way north of $500M to replace. So what. If we had - I dunno java (or maybe that fantastic London stock exchange .net code) we would still need the MF to process, with at least 5 9's reliability, the zillions of transactions per day of other people's money.

People love to whine about all the Apple stories. I would defy any of them to submit their own stories about all the other computer companies that are breaking new ground with this type of research.

Eh? This is hardly breaking new ground. IBM achieved a "retina display" 10 years ago. Kudos to Apple for identifying a supplier who could provide something similar but smaller for a device held closer to your eyes. But unlike Apple, IBM actually did the R&D themselves, doubling the state of the art at the time before bringing it to market.

Apple has become like the Microsoft of old. Repackaging old things and presenting them to an enamored audience of fanbois who oooh and aaah at all the wonderful things Apple/Microsoft "invented". I still meet people who think Microsoft invented the Internet just because they're ignorant of the rest of the tech market and were only exposed to Microsoft products. I'm starting to see the same thing happen with people who shroud themselves entirely within Jobs' reality distortion field.

People love to whine about all the Apple stories. I would defy any of them to submit their own stories about all the other computer companies that are breaking new ground with this type of research.

Eh? This is hardly breaking new ground. IBM achieved a "retina display" 10 years ago. Kudos to Apple for identifying a supplier who could provide something similar but smaller for a device held closer to your eyes. But unlike Apple, IBM actually did the R&D themselves, doubling the state of the art at the time before bringing it to market.

Apple has become like the Microsoft of old. Repackaging old things and presenting them to an enamored audience of fanbois who oooh and aaah at all the wonderful things Apple/Microsoft "invented". I still meet people who think Microsoft invented the Internet just because they're ignorant of the rest of the tech market and were only exposed to Microsoft products. I'm starting to see the same thing happen with people who shroud themselves entirely within Jobs' reality distortion field.

I don't fully understand your dislike of the usage of the word agile. Is it that agile == inexperienced/uninformed flash in the pan to you? Agile development is a widely accepted and practiced software development methodology. It has been for years now. Many (all?) of the big-wigs of the whole agile movement (if you want to call it a movement) are greybeards themselves. Do you think that no projects at IBM are developed using an agile methodology? I highly doubt it considering http://www-01.ibm.com/software/rational/agile/. Maybe I'm misinterpreting your intent since I'm not exactly sure what you meant by domain-specific meaning of agile, since clearly agile has well-defined meaning in the software development world. Are you saying that we should be trying to discuss alternate meanings of the word agile? E.g., 1. quick and well-coordinated in movement; lithe: an agile leap. 2. active; lively: an agile person. 3. marked by an ability to think quickly; mentally acute or aware: She's 95 and still very agile.

IBM seems to think that "validate input" is an appropriate term for this
http://www.ibm.com/developerworks/library/l-sp2.html

I'll go get the gun for you.

IBM recently announced LTFS (Long Term File System), which allows one to operate LTO-5 tapes as if they were a normal file system.

That's a very exciting technology which allows for the standardization of tape formats -- its specs are freely available in the LTO Consortium website and the implementation has been released under the GNU LGPL (see the LTFS website for links).

Tapes are not dead, certainly!

variable-length arrays were a surprise to me too. I had been looking for the quickest way to do an in-memory search to find keys so I could look up the associated pointer, and as we all know, c doesn't have variable-length arrays ... but c99 does. Use -std=c99 on the compiler line.

What this means is that you can treat your chunk of ram as contiguous, and realloc to grow it, and if you try to do an array access past the "old" bound, you no longer throw an exception. So your initial array allocation (in my case, 1024 elements) could grown (again, in my case, to a million) and still work.

Doing a bsearch on that is dirt quick. Being able to refer to any element, even those beyond the original length, using standard array notation, is just SO nice.

http://publib.boulder.ibm.com/infocenter/comphelp/v8v101/index.jsp?topic=/com.ibm.xlcpp8a.doc/language/ref/variable_length_arrays.htm

A variable length array, which is a C99 feature, is an array of automatic storage duration whose length is determined at run time.

If the size of the array is indicated by * instead of an expression, the variable length array is considered to be of unspecified size. Such arrays are considered complete types, but can only be used in declarations of function prototype scope.

A variable length array and a pointer to a variable length array are considered variably modified types. Declarations of variably modified types must be at either block scope or function prototype scope. Array objects declared with the extern storage class specifier cannot be of variable length array type. Array objects declared with the static storage class specifier can be a pointer to a variable length array, but not an actual variable length array. The identifiers declared with a variably modified type must be ordinary identifiers and therefore cannot be members of structures or unions. A variable length array cannot be initialized.

A variable length array can be the operand of a sizeof expression. In this case, the operand is evaluated at run time, and the size is neither an integer constant nor a constant expression, even though the size of each instance of a variable array does not change during its lifetime.

A variable length array can be used in a typedef expression. The typedef name will have only block scope. The length of the array is fixed when the typedef name is defined, not each time it is used.

Now don't tell me that isn't sweet! For some algorithms, it's a real game-changer, making them so much simpler to implement - and explain to others - that you never want to go back.

you didn't look to very damn hard... Oh, that was the point wasn't it.

http://www-03.ibm.com/linux/
http://www-03.ibm.com/linux/systems.html
http://www-03.ibm.com/linux/software.html

you didn't look to very damn hard... Oh, that was the point wasn't it.

http://www-03.ibm.com/linux/
http://www-03.ibm.com/linux/systems.html
http://www-03.ibm.com/linux/software.html

you didn't look to very damn hard... Oh, that was the point wasn't it.

http://www-03.ibm.com/linux/
http://www-03.ibm.com/linux/systems.html
http://www-03.ibm.com/linux/software.html

ETA: Currently 20 visualization options are available. Here's a Treemap based on the Nov. 2007 statistics: http://manyeyes.alphaworks.ibm.com/manyeyes/visualizations/supercomputer-center-treemap

ETA: Currently 20 visualization options are available. Here's a Treemap based on the Nov. 2007 statistics: http://manyeyes.alphaworks.ibm.com/manyeyes/visualizations/supercomputer-center-treemap

Be sure to check out IBM's Many Eyes visualizations (which includes a Treemap).

http://robocode.sourceforge.net/

Write Java or .NET code to destroy other bots! I haven't played/coded-for this game in several years, but I know somebody with the exact same problem as the Ask Slashdot poster... And the game is genuinely a LOT of fun, as there are lots of other bots freely-available to compete-against, some of which are pretty sophisticated (implement statistical targeting, a genetic algorithm, etc.).

See also IBM's introduction to it way back in 2002 -- which was around the same time a previous Slashdot article pointed me towards the game: http://www.ibm.com/developerworks/java/library/j-robocode/

No, they're not finding this out now, it's nothing new. For example, IBM offer a datamining service to do all sorts. Just search for "Social Network Analysis" and there's lots of stuff going on.

As a developer who works on (closed-source) enterprise software which runs on Linux (amongst other platforms) I'm nervous about Novell being sold. Though I develop on Fedora and primarily use RHEL for informal testing (we do formal testing on all the platforms we support) I'm glad that a solid, serious alternative to RHEL exists.

Obviously a sale of Novell doesn't necessarily imply any change for their Linux business (esp. as I understand it's one of their more profitable divisions) but it is likely (in the short term) to introduce some uncertainty.

The Linux market seems very healthy at the moment and I hope it continues to be at least a duopoly. Red Hat are a very cool company but I wouldn't like to see any company have a (virtual) monopoly in Enterprise Linux.

It is already over now, to put more data onto a magnetic HD you add more platters or increase the radius, that's all.

I believe I've found your dream hard drive:

http://www-03.ibm.com/ibm/history/exhibits/storage/storage_350.html

The best part is the 305-RAMAC CPU inherently provides that "smooth vacuum tube sound" when decoding mp3s...

yea it is.

and you should spare us your talking bullshit without things you dont know enough about.

after all, 10 do something 20 goto 10 can define basic in that respect.

what you are doing is pricky ignorant elitism. ironic that ibm doesnt think like you do.

http://www.ibm.com/developerworks/opensource/library/os-php-7oohabits/

Those boards tend to cost about a grand (they're not the only ones who say they have 'em) for a single cell and you don't even get all the PS3 hardware. Then again, you don't get access to all the PS3 hardware with Linux on PS3, either; the GPU, which contains about half of the system's power, is not accessible to you. Still, that's a ridiculous price by any standard. I can see paying no more than twice what a PS3 costs. You can now buy IBM blades with a pair of second-generation cell processors, the prices are almost certainly high, yet also almost certainly provide superior price:performance than buying a bunch of two-slot cell processors that you can install perhaps four of in a PC, tops. Hmm, okay, maybe not, this is IBM: It's ten thousand dollars for one box with two blades, that's four cells. They do have twice the double-precision performance, but that's still only four times the performance (in this context) as a pair of PS3s. So unless your job is less than embarrassingly parallel (i.e. you need access to the full data set, or no more than 32GB-overhead of it anyway, where this IBM box maxes out) you'll get better price-performance out of PS3s, even for double-precision fp math. But the federal government has never been shy about giving our money to IBM in the past...

Those boards tend to cost about a grand (they're not the only ones who say they have 'em) for a single cell and you don't even get all the PS3 hardware. Then again, you don't get access to all the PS3 hardware with Linux on PS3, either; the GPU, which contains about half of the system's power, is not accessible to you. Still, that's a ridiculous price by any standard. I can see paying no more than twice what a PS3 costs. You can now buy IBM blades with a pair of second-generation cell processors, the prices are almost certainly high, yet also almost certainly provide superior price:performance than buying a bunch of two-slot cell processors that you can install perhaps four of in a PC, tops. Hmm, okay, maybe not, this is IBM: It's ten thousand dollars for one box with two blades, that's four cells. They do have twice the double-precision performance, but that's still only four times the performance (in this context) as a pair of PS3s. So unless your job is less than embarrassingly parallel (i.e. you need access to the full data set, or no more than 32GB-overhead of it anyway, where this IBM box maxes out) you'll get better price-performance out of PS3s, even for double-precision fp math. But the federal government has never been shy about giving our money to IBM in the past...

Is anyone else here thinking: so what? Sounds like a press release with almost nothing of interest.

Here's the press release: http://www.ibm.com/news/th/en/2010/05/04/m796788v34229n07.html
Everything of interest was cut out by "Staff Writer"

If InfoGrok is going to just reprint a cut down press release, they should really say so.

to floating point, please, everyone should've read Everything you ever wanted to know about C types and part 2 (which explains fp too).

this will save a lot of time & questions to most beginning (and maybe mediocre) programmers.

to floating point, please, everyone should've read Everything you ever wanted to know about C types and part 2 (which explains fp too).

this will save a lot of time & questions to most beginning (and maybe mediocre) programmers.

cf. Anatomy of real-time Linux architectures and the Real-Time Linux Wiki

Not free or open source. I think IBM Tivoli Application Dependency Discovery Manager or TADDM (what a name) can help you discover the machines and dependencis you had there. http://www-01.ibm.com/software/tivoli/products/taddm/ But i don't know if it will help ou with your example of the UPS. And it will cost you the left nut and half of the other.

The Citrix stuff had little to do with it. Th Linux kernel developers favor code that is easy for them to integrate and maintain, and KVM fit better into that model than Xen. There are some situations where it performs quite a bit better too, and frankly few people care about those stuck with processors that don't have the right extensions to use KVM. Some good reading on the background here includes Discover the Linux Kernel Virtual Machine, Linux: KVM Paravirtualization, and The truth about KVM and Xen.