Slashdot Mirror

A lot of problems (though not all) would go away with the right GCC extension.

IBM DOES distribute Linux. Look here, for example. (That's just the first example I could find; I'm sure there are others.)

IBM cannot assert any of it's patents against linux, since they:

1. Contributed to the linux kernel
2. Would look very silly and incongruent for going against something it uses to make money

This is just counter-FUD to keep IBM's linux customers satisfied.

IBM has an article on Yellow Dog on PowerMacs. I personally can't see running anything but Mac OS X on a PowerMac, but to each his own.

• HP and Open Source
• IBM Open source projects
• IBM Wont Use Patents Against Linux
• Software giants feel open source pressure
• Sun, Java and Open Source
• Nokia fights Microsoft by addressing Series 60 developer complaints

I also have no doubt whatsoever that your statement is bald zealotry

If the current corporate adoption of OSS is what constitutes critical mass (ie a few marginal projects here and there), then continue to welcome our current microsoft overlords..

Verticle applications will be a different story, but there's very little that can run on LINUX that can't run on Solaris already. Never mind what gets written in Java.

I do think it's interesting that one of the features of the new version 10, aka the Open Source version, is that it will have the same SQL as DB2, so you can more easily migrate. Unfortunately, it seems that means in almost every case, reduced SQL Functionality.

Luckily, as a current user, I can't think of any instances where I was using the removed functionality. It speaks to how strong a product the existing Cloudscape is, though. Also, it would have been nice if they'd improved the DB2 SQL rather than trimming back Cloudscape.

The Cloudscape homepage: Cloudscape
And more details with links to PDF documents: Features and Benefits

I would guess that mysql would be faster for simple stuff, but Cloudscape could give it a run for it's money with support for more complex SQL.

Wouldn't know how it compares agains postgresql...

The Cloudscape homepage: Cloudscape
And more details with links to PDF documents: Features and Benefits

I would guess that mysql would be faster for simple stuff, but Cloudscape could give it a run for it's money with support for more complex SQL.

Wouldn't know how it compares agains postgresql...

However, generally when benchmarking one doesn't actually use both the Gbit NIC's (he's a double idiot if he failed to mention this). From the setup, he's got something silly going on. As a general rule, the NIC and the harddrive will need to use up roughly the same about of PCI bandwidth (they are writting roughly the same amount of data, plus or minus framing/headers for the frames/sectors). 140Mbit/s * 2 = 280Mbit per second. So your telling me the "other misc" peripherals are using up 2-3 times the PCI bandwidth of a Gbit card and a harddrive running flat out? What else do you have hooked up to your machine?

More likely, he's running out of CPU, interrupts (they are queuing up faster then they can be serviced), or scalability in the kernel somewhere else, long, long before he's running out of PCI bandwidth. You took the two biggest PCI bandwidth hogs on his machine and showed they used up possibly a fourth, (maybe even a third if you throw in lots of overhead) of the PCI bandwidth, and then claim the "extra stuff", takes up the rest. I'm not buying it.

I'd hear what your saying, if it wasn't just silly. You can't even get close to the Gigabit thru put if you don't have a 800Mhz machine (he doesn't say how fast his machines are). I'm not sure if that uses up all the CPU power of the machine or not.

I've had exactly the problem he describes over NFS with 2.4Ghz Xeon's machines with Gigabit cards on the Northbridge. In the end, it was that NFS protocol has latency in it. In Ethernet parlance, the sliding window is too small. So while I could easily in aggregate get a Gigabit of bandwidth in a number of connections, I could never get a gigabit of bandwidth in a single connection. I wouldn't be shocked if this is also the case with Samba.

I could put about 500Mbit/sec over the cards via NPtcp, but for a standard file transfer over NFS, I couldn't exceed about 14-15Mbyte/sec, which is ~ 120Mbit/sec. You want to get the harddrives, and the PCI contention out of the loop to see what the problem is. Run NPtcp on both ends, check what the maximum thru put is. That'll give you a good number to start with to see how much data your machine can actually push over the wire. From there, it's easier to start creating artificial contention and CPU usage to see what could be causing the problems. If you believe it to be the PCI bus, you can fiddle with the timings of the PCI bus to enhance the bus utilization. Read up on it here: http://www-106.ibm.com/developerworks/linux/librar y/l-hw2.html

It's down at the bottom.

Kirby

IBM.

Large company.
Large bureaucracy.

Brilliant engineers.
Large bureaucracy.

What a small, agile company headed by a techno-geek would do is not necessarily the same as a lumbering giant headed by a sales/marketting critter.

The only reason that the red-tape may be circumvented to do something smark is that the CEO (http://www.ibm.com/ibm/sjp/bio.shtml) was instrumental in getting Linux into IBM. But it's a lot of red tape for one person in such a large bureaucracy.

Informative.

A GRID is a non-centralized distributed system, sharing storage, processing and connectivity with quality of service guarantees, over open protocolls. Is that about right?

Last summer, I was in Barcelona for a class trip. We visited the European Center for Parallellism of Barcelona (Like there are redundant arrays of Barcelonas. Well, there's the smaller town/city of Badelona right outside Barcelona.)

Anyway...

The CEPBA is a partnership with IBM. I talked to one of the IBM researchers there. They very much believed in the GRID as a distributed system that would hold all your personal data for you, and where you would buy your processsing power and storage from big vendors. Thus users wouldn't have to keep up with the hardware advances themselves.
I suggested that some of the less time-critical processing and storage resources could be provided by the nodes themselves. He did not buy into this idea.
This philosophy reminds me of the old time-sharing systems of yore. If GRIDs are really deployed to a vast number of users, I would predict that users with smaller processing power would team up and sell their (lower quality?) processing to other users, much as the F***** Article says. The only thing that could preclude this would be an enforced requirement that only reputable vendors provide services with QoS guarantees, and digital restrictions stop users from selling their cycles.
Why not have several layers/levels of quality and security? That's much of the point of QuS, isn't it -- telling what's your worst offer and the client deciding beforehand wether to accept your service level?

Oh, look: Now their saying that institutions really WILL be able to pool their resources What is GRID compiting
Looks like I was "getting with the program" better than that IBM guy was. Either he misunderstood or given wrong information (something I've done a few times and been ashamed of), or is a firm believer in that Those darn kids will never provide the computing power! Only me and my VAX!

Of the 360-based operating systems, IBM's TPF has a major presence in the airline industry, but this probably isn't the system in question. TPF tends to handle ticketing and reservations. TPF stands for the Transaction Processing Facility; it's the descendant of the old Airline Control Program (ACP) developed for Sabre. Sabre in fact is still running TPF, although I believe they're busy transitioning away from the mainframe to Tandem's er I mean Compaq's er I mean HP's NonStop/UX.

Of course, it might not be an IBM mainframe at all; Unisys has a niche in the airline industry. But heck; given that this is route planning, just about anything from AIX to z/OS is a possibility. Even *shudder* Windows.

Within the google search I fonud Developing LSB-certified applications http://www-106.ibm.com/developerworks/linux/librar y/l-lsb.html maybe this is the subject we should be discussing, how many applications ARE following LSB guidelines?

Also, why is the Linux Standard base hosted by the open group? Isn't the Open Group doing it's part already? so the other way arround becomes less relevant. ;)

http://www.opengroup.org/lsb/cert/cert_prodlist.tp l?CALLER=index.tpl

That's funny. This is what Google found me, as I also rushed to check the marvel laptop :) Note however that this configuration is 1.4 GHz instead of 1.7 and a 40 GB HDD instead of a 60 GB one.

The ThinkPad T41 is currently priced at $2,522.13 . They couldn't find another laptop that is more cost effective than that? One of the benefits of Linux is first the OS is free, but also it doesn't require the Spartan hardware of Windows. For $2,522.13 I could simple get one of these and not worry about getting sound drivers etc. to work.

There is one thing a software raid can't do - hot-swap. A must have feature if you consider doing backups to hdd

You're on crack. I've been hot-swapping SCSI and IDE under Linux for ages. IDE hotswap is unofficial and your vendor won't talk to you no more if he finds you've been using hdparm -U and -R, but SCSI hotswap is just part of the norm. Hell even IBM has an article detailing the /proc command: Here. Bascially

echo "scsi remove-single-device host channel ID LUN" > /proc/scsi/scsi"
echo "scsi add-single-device host channel ID LUN" > /proc/scsi/scsi"

Catia.

A quote from IBM: "For each of the past 11 years (1993 - 2003), IBM has been granted more U.S. patents than any other company. During that period IBM has received 25,772 US patents. In 2003, IBM received 3,415 U.S. patents, breaking the record it set previously for the most US patents received in a single year."

Gotta admit thats kinda impressive...

Microsoft may want to earn more respect now that they have started to share their $60+ billion war chest with their stock holders. Fair enough. But they can't earn my respect my just saying that they did 2,000 patents last year, and may do 3,000 this year -- so what? Lets see some sustained performance or at least publish their sustained historical performance...

The question is can they deliver patents over the long haul... they already got the easy ones... Patent No. 6,748,582 (Microsoft's patents the "to-do list").

I am forgetful but not yet impr,eTOEd...

Two...I have two words for you!

Seriously, I don't see the problem, so long as companies like IBM and (dare I say it) Microsoft continue to do research in this area. That is the real value of companies that are committed to *real* research in revolutionary sciences and technology.

Of course, US companies don't have a hammerlock on this research. There is a lot of work being done internationally in the area, by corporations, and by educational/research institutions.

---anactofgod---

Actually, there is a fascinating research project by IBM, the History Flow tool, which charts the history of a Wikipedia article. Email conversations with the guy in charge have revealed that they are going through the (internal, corporate) motions required to get it released for use, and it may possibly even end up GPLed or something like that.

OS/2 of course!

The Win32 threading and synchronisation models are ridiculously powerful compared to *nix, which is precisely what makes it so hard to port a lot of Win32-based software to other platforms. The fact that you can't do a simple operation like "wait for a mutex to be released or a socket to become readable" deserves to be a joke about legacy operating systems, not a persistent reality. At least BSD's kqueue comes close.

If that is true, then it's a shame that the performance of the Win32 sockets are so meagre compared to the Linux implementation. Take a look at this article on Developerworks. Maybe you can spot the changes required to close the performance gap between Windows and Linux (Linux running about 2 and a half times faster on the same machine).

And I think I'll take you to task for your blind assertion that "you can't do a simple operation like "wait for a mutex to be released or a socket to become readable" on a Unix platform. If you call pthread_mutex_lock in 'fast' mode it simply waits for the mutex to be released and will pick up as soon as the mutex becomes available. And there are plenty of other options around. It's also totally trivial to write a spin-check to check the TCP status of a socket.

Cheers,
Toby Haynes

For crying out loud, where have you been so long? Quick, let them know before it's too late.

OS/400 (the OS of the AS/400, no, sorry, the iSeries, no, sorry, the i5) isn't much like any other operating system I know of. All the unicies and Windows all feel like variations on a theme in comparison.

OS/400 (the OS of the AS/400, no, sorry, the iSeries, no, sorry, the i5) isn't much like any other operating system I know of. All the unicies and Windows all feel like variations on a theme in comparison.

This link seems to suggest otherwise.

I guess that is a matter of opinion. Here is their software patents. You be the judge.

We test using the following on web apps:

• Target browsers for intranet apps (even though we use standards as much as practically possible)
• W3C validators for HTML, CSS, and Links
• Validators within WebSphere Studio (Java, JSP, HTML), HomeSite (HTML) and TopStyle (CSS)
• JavaScript Console and Debugger in Mozilla/Firefox
• JUnit
• Cactus
• People. The users. The project owners. Us. Other web developers on e-mail lists.

• We aren't currently using an automated tool to test the front-end flow, because we haven't found any good, easy-to-use, and cheap tools that support a modern version of DOM/JavaScript usage. If you know of something that you like and works, I'd love to know about it. I've tried httpUnit, but had trouble setting it up and it didn't support all the DOM methods we were using at the time.

True, but I doubt that a multi-core chip will be on par with a similar dual-cpu setup, you still need to get the heat away from that single cpu. It's very possible you will only get about the same 15-30% boost in speed you get from HT.

Your doubts are baseless. IBM's Power4 CPUs, which have been out a couple of years, don't have any such problem.

If heat becomes enough of an issue, exotic (for a PC/workstation) solutions exist already. For example, water cooling has been an established solution in mainframes for decades before PC enthusiasts adopted it.

From what I understand multi-core designs have all cores on a single piece of silicon at the center of the CPU just like uni-core CPUs.

Yes, it's a single piece of silicon with two separate CPUs on it.

The article misses a fundamental point about dual core CPUs. IBM calls a 32 CPU p690 a 32 CPU machine. It is actually a 8 CPU, 4 core per CPU as per the article definition. i.e. 32Check http://www-1.ibm.com/servers/eserver/pseries/hardw are/whitepapers/power4.html for more details.Quote "Four POWER4 chips can be packaged on a single module to form an 8-way SMP. Four such modules can be interconnected to form a 32-way SMP." Sun on the other hand calls a 144 CPU E25K a 72 CPU machine. It is actually a 72 CPU, 2 core per CPU as per the article definition. Quote "Based on two UltraSPARC III pipelines". Check http://www.sun.com/processors/UltraSPARC-IV/ for the detail. It will be interesting to see what HP deliver for a dual core CPU this year. Oracle must love the Sun E25K as they charge Enterprise licences based upon the size of machine.

True, but I doubt that a multi-core chip will be on par with a similar dual-cpu setup, you still need to get the heat away from that single cpu. It's very possible you will only get about the same 15-30% boost in speed you get from HT.

From what I understand multi-core designs have all cores on a single piece of silicon at the center of the CPU just like uni-core CPUs.

Obviously, you probably think a Mac, SGI, or the ever popular for OpenGL pSeries is cheap as well, given the price bracket of the cards to run this beast, $300-400+. With the realistic low end at the 7500/Mobility FGL 7800(The only chip solely in a non Apoo or jobstealing laptop)- and the highend (read: always out of reach of mere mortals due to the oligopoly created by the buyouts) is around the x800.

But you arent forced to do X, buy X or involve yourself with X

Well, that doesnt justify your point anymore, since some of us dont have shitloads of money.

To allow anything lesser means a sacrifice in innovation

This one is probably from the gaming community that isnt old enough to remember when one had to actually made your equipment last for a long time (when a usable PC still cost $2000) by getting every inch of that hardware used before thinking of that new upgrade. I'd think innovation would be to step up to the plate and get all the performance/quality out of the lowend/midrange before those who just blindly buy whatever is topend for $500+ expecting that everyone has their resources at hand. I'm not exactly asking for one to go down to Virges or integrated, but asking for innovation to happen within the hardware that's in the low-midrange first (read: Radeon 7500-9600 non LE).



(No nvidia quotes were given due to buyout of 3dfx)

The orient, anyway-- there's also some evidence China may have been a source. A website which discusses the origins can be found here, and at least sounds plausible, but whose accuracy I am unqualified to judge. However, notable (and something I remember from may days with the SCA as being accurate) is that the exact pieces, moves, and board played on have varied over the ages. The game we commonly call chess today looks notably different from the Indian original (enough to send Deep Blue into a fit if you set it up for a match), and is a composite from several influences, having changed markedly over the centuries. (There still are a lot of obscure variants out there.)

So, while you're correct that this may have been the church altering history, it may be that life imitates art. This alteration of history might be why the CURRENT European style pieces have the CURRENT moves. You'd really want to ask an expert about it-- or at least, to recommend a good book on the subject. But more likely, it's unprovable at this point.

Woo hoo! DLLs! What won't these brilliant folks at Microsoft think of next? Wow! A whole library of executable code! Shared between multiple processes!....

Forgive me if I'm underwhelmed. Can ASP.NET run on one of these or one of these?

So you want fast, and you run on commodity/toy computers? And you're vendor-locked into doing so?

Also seems like everyone is complaining about ASP. ASP and ASP.NET are two completely different beasts. ASP was buggy and a pain in the rear to work with. ASP.NET, however, was amazingly simple to use with an amazing debugger (VS.NET). Please keep on the subject and leave out ASP.

I note with interest that you did not say that ASP.NET is not buggy.... :-)

Ask these guys.

If you want to build an IDE for PHP, you could do worse than build something on top of Eclipse. It's not just a Java tool, it's been done for Python, and the plugin architecture is pretty sweet.

Havent seen much IBM Power4 and Power5 boxes, have you?

Linux on POWER

Whatever happened to the IBM Roentgen?

IBM article about the IBM Roentgen That was LCD technology, and the big deal at the time was they were first to get to 200dpi (previous record was the Monet at 157dpi). It's 2,560 x 2,048 pixels and 16.3" diagonally, dot pitch of around 0.127mm. At one point, it was supposed to end up in ThinkPad laptops. (Wish I had one for my laptop!)

However, I don't know that anyone sells it (can't find anything).

Other alternative:

ViewSonic Vp2290b 22.2 in LCD Monitor
3840 x 2400 Pixels - Dot Pitch 0.125 Mm - VGA, SXGA - Sku: vp2290b
Price Range: $5,699.00 to $7,046.04 at 10 stores

The ViewSonic VP2290B/VP2290B2 are 203ppi LCDs.

The era of the big IT company may well be at an ending due Free & open source. I think the company's will be the one's like Mandrake!

It doesn't appear that F/LOSS is hurting the biggest IT company at all.

There are alternatives. Sure, you might not be able to get them brand new for reasonable prices, but that and the higher end of the older RS/6000 series(yep, they have 64bit models, and they arent just limited to the lowerend powerpc processors) do bring a decent alternative that is affordable (and can run opensource software if you have to ask). The only thing you wont get is chrome, but you'll get a well-built true to UNIX machine with many options, and rock-solid stability.

Am I the only person here who thinks that an iMac G5 is no good?
Well, when you can get a POWER based system without the chrome tax and true UNIX, the G5 ends up being no good. If anything, the people who clamor for a lowend headless POWER3/POWER4 system ought to look this direction, since they've been around for ages. The only things you'll leave behind are the zealots and the Apple chrome case.

Millipede

IBM recently announced *nix servers based on Power5, which makes me wonder if some of the resource going towards G5s got diverted to P5s. They're certainly not the same chip, but the similar foundation probably means they're produced in the same plant.

BTO always, always takes longer, but I doubt that's to blame this time.

It's their fault - difficult to sell a computer when your supplier can't make the required CPUs. Hell, Apple just took the iMac off the market until September because of G5 delays.

Like the other A/C said, when it's done, it'll be shipped to you directly from Taiwan.

It's very trivial in our environment. We just send AppWorx a large check every year, then keep two or three contractors busy tending it. It handles all our needs, except those met by MQSeries. But hey, we're a twenty billion dollar business. Your mileage may vary.

The fact is people like to be able to archive data in a tangible format, but they will demand increased efficiency, especially as file sizes increase over time.

Hence Holographic Data Storage, dramatically increasing the data per square, or in this case cubic inch of medium, while possibly (as technology improves) still providing an inexpensive easy to store medium.

And quite frankly, I think our new robot overlords wouldn't settle for anything less.

As data storage/volume ratio increases, so does the throughput of the memory BUS that is the human race.... Gotta love all that legacy hardware.

Best thing is to email IBM and Novell to sponsor that or phone them. I mean the nearest to entertainment IBM go to in sponsoring was some pretty mediocre animations. *cringe*

Lets put that in perspective for a moment. You don't mind the government putting your reading habits in a database, which can be used to gather statistics and flag books. If a person reads too many "red" books, they go up a few marks on the terrorist charts.

And just what is this going to mean for AIX Admins and programmers?

Great. Another America hater who's ignoring American companies that support and defend Open Source, American companies whose name is n with Linux and free downloads of Linux, and an American company who has been mailing free of charge 10 gigs of their Open Source software. Yeah buddy, it's all of those bad old American companies.