Microsoft Finally up for Distributed Computing?

ReeprFlame writes "eWeek has reported overhearing Microsoft's plans to finally get into the distributed computing market. Considering that the Windows platform has never had the ability to parallel compute in the past, it leaves great potential to the company's operating system development. From current *nix systems we have today, such a grid proves very useful, especially in the serving arena. However, we are unsure of Microsoft's target for the software. Would it be an addition to home users computers as well as the server versions of Windows? As of now it is unclear, but Microsoft probably will bring this situation to life in the near future since it does hold alot of power for them over other platforms."

Comments

Oh great...

[Anonymous Coward]

now we have to worry about the blue wall of death.

Re:Oh great...

[niteice]

I thought it was the many simultaneous BSODs around the world.

Wait, those are from the people who upgraded to Windows 98 the day it was released.

Re: Oh great...

[Black Parrot]

> now we have to worry about the blue wall of death.

No, the idea is to use one node as a dedicated BSOD server, so the rest can stay up all the time.

Re:Oh great...

[sentientbeing]

I thought Windows already supported distributed computing straight out of the box?

I heard the windows equivalent to Beowulf clusters were called 'Botnets'.

Re:Oh great...

[Clevershutter]

Windows is not a network OS and it has virtually no security, which is critical for Grids. Is MS going to adopt Linux to do this?

Re: Oh great...

[cas2000]

> No, the idea is to use one node as a dedicate
> BSOD server, so the rest can stay up all the
> time.

what's the point of that? wouldn't it be better to have a BSOD server which generated BSOD events for other nodes, to save them the trouble?

I know...

[frickenhell]

They'll be secretly using your CPU cycles to compile their latest version of Windows.

Re:I know...

[basvdlei]

They'll be secretly using your CPU cycles to compile their latest version of Windows.

Just in... Thanks to all this new Big(TM) technology Longhorn will be released next month!

Re:I know...

[ultranova]

I doubt it, not on the basis on their morals or anything like that, but how much processing power do you really need to compile this:

Apparently more than my poor 1GHz Duron has available, since gcc refused to compile it.

But then again, even if it had compiled, where would I had uploaded it, now that SuprNova is down ?

And are you sure this code is the source for Longhorn, and not the leaked WinNT source ?

Re:I know...

[Comatose51]

Did you include the and ? Obviously you got the joke and the moderator didn't...

Re:I know...

[jack_csk]

Actually, I am worrying if the leaked NT source could be used against ReactOS in the future. Nobody can guarantee whether Microsoft will copy SCO's trick if ReactOS becomes a threat to them.

Clusters

[rastakid]

From current *nix systems we have today, such a grid proves very useful, especially in the serving arena.

Keep in mind though that Windows clusters are existing. Of course this is not the same, but it's not like all servers are single-machines.

Third party solutions got there first

[Anonymous Coward]

Don't the spammers and virus writers have this technology already in their botnets?

I guess Microsoft is imagining a Be-- stop! put down that bat!

Not a good idea

[scsirob]

Windows has proven time and again to be designed for stand-alone situation. All network and security add-ons have shown to be just that; add-ons..

Distributed computing simply isn't part of the base design. Morphing Windows into something it isn't will once again be a task for their marketing department, not engineering.

Re:Not a good idea

[Sliptwixt]

Distributed computing simply isn't part of the base design Was it part of the base design of Linux?

Re:Not a good idea

[tesmako]

The NT kernel is modular and well-design, I can't imagine that it would be more troublesome than it was for Linux (and probably significantly less work).

I have no idea what you are talking about when it comes to networking and security in NT, sure the WIN32 part is troublesome to keep secure, but NT in itself has no such problems.

Re:Not a good idea

[tomhudson]

Security and networking weren't part of the base design of Windows OR its predecessor,Dos, unlike all the *nixish operating systems, which were designed from the ground up with the intent of running multiple processes with multiple users.

Sow's ears and silk purses, etc.

Re:Not a good idea

[peragrin]

The NT kernel is well designed.

The rest of the OS has way to much backwards compatibility to be able to strip things out.

Linux can run on clusters because you can install only those chunks that you need. in Windows every processor would also have to run the entire GUI. Even if it is never used.

Why do you think Longhorn is getting a full command line shell setup?

Re:Not a good idea

[oexeo]

The NT kernel is modular and well-design, I can't imagine that it would be more troublesome than it was for Linux (and probably significantly less work).

I have no idea what you are talking about when it comes to networking and security in NT, sure the WIN32 part is troublesome to keep secure, but NT in itself has no such problems.

Flamebait? Please mods, this is the reason slashdot is losing readership. It's difficult to have a decent discussion, when all opposing views to the group opinion (right or wrong) are essentially censored down to obscurity.

Re:Not a good idea

[Gumber]

This is a load of horseshit.

First: The WindowsNT line (WinNT, Win2k, Win2k3 and WinXP) isn't descended from DOS.

Second: WindowsNT had a rich file and process permissions and auditing model baked in at a low level that exceeded (and may still exceed) what Linux has today. The problem is that the default OS config was and is relatively permissive.

Re:Not a good idea

[Tim C]

Exactly what I was going to say. There's absolutely nothing in that post that constitutes flamebait.

This place is really going downhill; time was, you could have a reasoned, rational discussion. That's getting harder and harder all the time these days, unfortunately.

Re:Not a good idea

[tomhudson]

This is a load of horseshit.

First: The WindowsNT line (WinNT, Win2k, Win2k3 and WinXP) isn't descended from DOS.
</ecode.
No, your response is knee-jerk horseshit. Learn to read what I wrote:
<blockquote>
Security and networking weren't part of the base design of Windows OR its predecessor,Dos

Windows was never just NT, despite Microsoft's wishful thinking tha we should forget their flawed past.

What the fuck do you think they had to call it "NT" for - it was supposed to mean "New Technology", another attempt to get it right, and distance themselves from their previous efforts -

1. the Windows286 and Windows386 runtime environments that were linked directly into the application and provided a windowing environment for your DOS app (for example, Ami Pro way back when ...)
2. later on, the Windows 3x series, again, running over DOS
3. then the Windows 9x series, this time with DOS built in.

There was also the side trip with OS/2 and Presentation Manager (OS/2 was a collaboration between IBM and Microsoft that Microsoft sabotaged by insisting that all code be in assembler, while it was developing Windows 3.0 in C on the side).

Re:Not a good idea

[adeydas]

but somehow that is going to change according to the beta testers of windows longhorne. they say that its a monster and is much better that linux.

Re:Not a good idea

[westlake]

Windows was never just NT, despite Microsoft's wishful thinking tha we should forget their flawed past.

there is nothing to be gained by flogging a dead horse.
ignoring the odd embedded system, the O/S line that began with DOS is extinct.

Re:Not a good idea

[abradsn]

Why isn't Linux on every desktop then? Maybe windows has evolved over the decades.

Re:Not a good idea

[tomhudson]

First, there are LOTS of win3x and win9x systems still running, so it's not exactly a dead hrse.

Second, you're the one who equated windows with winnt to the exclusion of all else, not me.

Third, those who ignore the past are doomed not to recognize it the next iteration - when BlowHorn comes out, etc ...

IF Microsoft ever comes out with it (target date the end of the decade), it will already be obsolete.

Re:Not a good idea

[spectecjr]

Security and networking weren't part of the base design of Windows OR its predecessor,Dos, unlike all the *nixish operating systems, which were designed from the ground up with the intent of running multiple processes with multiple users.

Given that you're the one claiming that we shouldn't ignore the history of Windows when people pull you up on the stupidity of this comment, let me show you that you shouldn't ignore the history of Unix, which also WASN'T designed from the ground up with networking and security in mind.

From Dennis Ritchie:

"The first fact to face is that UNIX was not developed with security, in any realistic sense, in mind; this fact alone guarantees a vast number of holes."

Unix started out life with 14 character filenames.
No security.
No networking.

It wasn't until BSD that you had filenames up to 255 characters.

Learn some freakin' history.

Re:Not a good idea

[Bas_Wijnen]

I don't understand why you are posting as AC... If you're just reading anyway, you can't be too concerned about your karma? And I have a hard time seeing that post being modded down (except as off topic, but that's usually given to posts which have nothing to do with the thread, not the article, and yours is quite on topic for the thread).

Re:Not a good idea

[gad_zuki!]

What? The loud mouth pro-linux kiddies whose windows expereince is limited to videogames have no idea whats going on in the enterprise? On top of their FUD they knock down informative windows posts in defense of their pet OS? Say it aint so!

sigh, if there is a reason why "slashdot is losing its readership" (is it?) its because of evangelical mods and bullshit anti-MS posts.

If they want Open Source news they can visit NewsForge and stop whining about windows and other commercial apps.

Re:Not a good idea

[tomhudson]

Why isn't Linux on every desktop then? Maybe windows has evolved over the decades.

Linux at 10 was way ahead of Windows from Microsoft at age 10. http://www.computerhope.com/history/windows.htm 3 years in development + 7 years after its' introduction, Microsoft was just introducing Windows 3.0. I remember buying that piece of shit. The ONLY good thing about it was that it was easy to uninstall.

So, which would you rather have as a desktop - Windows 3.0 or last year's linux distro? More relevantly, who is going to be further ahead in the NEXT 10 years?

Microsoft has lost momentum, and is now reduced to trying to play catch-up in terms of features.

In terms of ease of install, and ease of maintenance, and ease of updating, linux distros win (just did an upgrade at the office from SuSE 9.1 to 9.2 Friday - almost 6 gigs of software brought up-to-date painlessly ... impossible with Windows where every service pack and update is feared for what it will break).

The problems with Windows are three-fold

1. the core has too much crud that was grafted into it that seemed like a good idea at the time (performance, etc) but shows its age ... there's some really OLD stuff in there that can't be rewritten without breaking too many other things. Their designers have admitted as much. Look at the way that numerous Microsoft products use undisclosed windows api calls, then they can't modify/fix something related because those calls, while not part of the public interface, have to be maintained because removing them will cause their software to puke. They end up having to write more code to handle more special cases ... over and over and over

   Linux, on the other hand, has no undisclosed api.

2. Windows target market is to be all things to all people - Gates himself has said "I only want our fair share of the market - and that fair share is 100%". In trying to be everything, it is neither fish nor fowl.

   Linux, on the other hand, is a kernel. It's agnostic in the sense that it is up to YOU as a user/developer/distributor to do what you want with it.

3. Reputation - Microsoft and Windows both have a terrible reputation, in terms of public relations (convicted monopolist, all-grasping with their "passport", etc...) and technology (everyone knows what the "Blue Screen of Death" means).

   The BSDs and Linux have a much better reputation. People don't use Windows because they want to, but because they have to. But that's changing.

Look at how long its' taken Windows to get to where it is today (21 years + 3 in development = 24 years). Linux is half that. Wanna bet that 10 years from now Windows is still on almost every desktop? I doubt it.

Re:Not a good idea

[tomhudson]

Why do you think I said *nixish and not unix. The parent OS was MULTICS, which had selective file sharing, access controls, etc.

access control
The Multics feature that checks if a user can do something. The user identity, established at login, is checked against the ACL of the thing being accessed.

[BSG] User access to segments is enforced by the hardware in bits in the SDW (see REWPUG). Segment control, which keeps track of all of the processes having SDWs for a segment (via a database called the system trailer segment, str_seg) is equipped to revoke access to a segment instantly when the ACL of a segment is changed. The connect mechanism assures that in spite of the associative memories in which access can be cached, access can be revoked in mid-instruction (see EIS) if need be.

and

ACL
Access Control List. A list of person names and project names in combination specifying individual and group access rights to a segment. First introduced in Multics.

The ACL on a file might look like
rew VanVleck.SysAdmin.*
r Backup.SysDaemon.*
rw *.SysAdmin.*
This ACL has three entries. Eacn entry specifies the access permission, or mode, for a set of users.

The modes for segments are
r read
e execute
w write
null no access

The modes for directories are
s status
m modify
a append
null no access

Each ACL entry also specifies the Person, Project, and instance tag that selects a set of users, either as an explicit value or as "*" meaning any. Instance tag describes how the user identity was authenticated: "a" means an interactive login, "z" means a daemon, "m" means an absentee job.

The ACL is always sorted, most specific entries first, and the first entry that matches determines access. That is, there are not "deny" entries as well as "permit" entries, which makes searching the ACL for a match fast and simple compared to Posix or Windows NT.

Another Multics design choice is that each user has a single group identifier, and groups cannot contain other groups, so that the matching on group is easy: check for *, check for string match.

Each directory contains two initial ACLs, one for segments created in the directory, the other for directories created in the directory.

See also extended access and mandatory access control.

So, as I said, *nixish systems (not unix, but unix-like systems) were designed with security and multiple users in mind :-)

Re:Not a good idea

[spectecjr]

Why do you think I said *nixish and not unix. The parent OS was MULTICS, which had selective file sharing, access controls, etc.

You said all the *nixish systems. That includes Unix.

Re:Not a good idea

[tomhudson]

Last I looked, even Unix (which is NOT a *nixish system - it's not unix-like, it's unix) has MULTICS as its' parent.

A quick analogy: I am not "like" me - I AM me. The difference is between resemblance and identity. Another example, stretching it a bit:

"Is your dog dangerous"
"Nope"
... goes to pat dog, gets bit ...
"Hey! You said your dog wasn't dangerous!"
"That's not my dog."

Re:Not a good idea

[suffe]

No, no no. Quiet! You are ruining a perfectly good carma increaser post that has been used at slashdot since the dawn of time. Slashdot, where posting the same thing over and over again acctualy makes sence.

Re:Not a good idea

[tomhudson]

Actually more. Just that some ideas withstand the test of time better, I guess.

Who knows, maybe somebody somewhere is making the next great thing even as we sit here.

Re:Not a good idea

[tomhudson]

Perhaps if you're going to have such a personal definition of "*nixish" you should define it before you use it rather than after you get called on it.

It's not My definition.

What do you think "linux" is - it's not Unix, that's for sure. Do a google for "linux is not unix" (include the quotes). Easy enough to remember - linux - Linux Is Not UniX

And "GNU" - "Gnu's Not Unix"

It's people like the litigious bastards who would like to think that you can dump millions of lines of unix code into a *nixish system such as linux. But you can't. Linux is not Unix, even if they share some common functionality.

You can also google for the term "*nixish" and find over 1,500 references to it being used the way I use it - "unix-like" operating system, as opposed to "unix" itself.

Or you could see what the wiki has to say:

Unix-like
From Wikipedia, the free encyclopedia.

A Unix-like operating system is one that behaves in a manner similar to the UNIX system, while not necessarily conforming to or being certified to any version of the Single UNIX Specification.

...

Early Unix-like systems

The first "Unix-like" operating systems were developed because of AT&T's licensing of Unix, which prevented the sale of Unix to commercial organisations. The Unix-like operating systems that were available in the 1980s and early 1990s included Idris, Coherent, UniFlex and Minix (a computer science teaching system).

So, *nixish systems have been sold for a couple of decades ... it's not something I just made up off the top of my head to justify my argument.

Re:Not a good idea

[Andrew Cady]

Linux at 10 was way ahead of Windows from Microsoft at age 10. http://www.computerhope.com/history/windows.htm 3 years in development + 7 years after its' introduction, Microsoft was just introducing Windows 3.0. I remember buying that piece of shit. The ONLY good thing about it was that it was easy to uninstall.

While this is true, the reason for Linux's quick progress was merely superior design; also, Linux could not have been made without the GNU components which were started some time before 1984, when RMS posted the GNU Manifesto.

Microsoft's operating systems have been low-quality because they were forced to maintain compatibility with their legacy systems, which had relatively shoddy design (indeed, this compatibility was their only selling point). Linux did not have this burden; it needed only source-level compatibility with Unix (still an unfortunate burden, but much less significant). Under Microsoft's same burden (though also starting from scratch), WINE has been consistently behind Microsoft. ReactOS may materialize, but it is not progressing with impressive speed -- although the comparison is probably unfair, because not much effort is going into it, relatively speaking.

I do not use proprietary software personally, except (when I have no control over the system) to use putty or ssh to connect to other systems, but I have seen Windows XP, and read something of its design, and there is no doubting it: Windows is catching up. In five years, I imagine Windows will be the technical equal of GNU/Linux. It is already superior in a number of ways. It will still cost more, which will be Linux's selling point; and it will still be proprietary, which is why I will not use it. But I think it is self-deluded to suppose that Microsoft cannot produce as technically good a product as free software, now that it has finally got over its backwards-compatibility burden.

There is an interesting parallel, I think, between MS Office vs. OpenOffice and Linux vs. XP. MS Office is obviously superior, yet OpenOffice is nevertheless good enough, and progressing faster owing simply to the fact that it has more room to improve. Meanwhile, Linux is somewhat superior to XP, but XP is also quite clearly good enough. Frankly, there's not much further to go on either side. A point of rapidly diminishing returns has been reached (with XP for Windows, and many versions ago for Linux). Application support is the vital factor now.

I feel I should also mention that Linux is not as superior as many would like to think. Plan9 is an obviously superior Unix (which, unlike GNU, has a valid claim at not being Unix), and EROS demonstrates the folly of any Unix-like design at all. Mac OS X also deserves a mention -- it is on the whole technically better than Linux, though less so than Plan9. And as far as implementation goes, each of the three free BSD's have bested Linux on a number of fronts. In particular, the free BSD kernels were always very stable, whereas Linux, for many years, was stable only in comparison to Windows.

Windows clusters don't make sense

[Xpilot]

Windows is an overly-bloated OS which is very GUI-oriented and is not modular or flexible for cluster node usage. Processing nodes usually don't even have a monitor or keyboard, much less a GUI and a mouse. Windows isn't much use there. Nor can you strip out the parts you don't need, or customize the kernel for performance. Plus, Microsoft's incredibly expensive and anal licensing makes a Windows cluster not worth the effort or money. I mean, Linux's licensing cost is 0, and 0 scales infinitely ;)

Say what you like about Linux "not ready for the desktop", but Linux (and *nix in general) totally rules the clustering arena.

Re:Windows clusters don't make sense

[heffrey]

Why do you need a monitor, keyboard and mouse to run Windows? It works perfectly well with no GUI through remote terminal access.

Do you know anything about Windows?

Re:Windows clusters don't make sense

[FooBarWidget]

In other words: Windows is not ready for the cluster!

Re:Windows clusters don't make sense

[ergo98]

While you're probably aware of grid computing from a perspective of "huge server farm at research organization X", I think the more practical use is corporations that often have 10s of thousands of extremely powerful workstations. These PCs are extraordinarily underused, and if there was some secure, reliable method of distribution processing across them (transaction calculations, actuarial processing, whatever) then that would be extremely valuable.

Re:Windows clusters don't make sense

[Anonymous Coward]

Processing nodes usually don't even have a monitor or keyboard, much less a GUI and a mouse. Windows isn't much use there.

We have some 2500 Windows servers where I work. None of them have monitors, keyboards, or mice. If we need a KVM it's typically to get into the BIOS, not the operating system.

Nor can you strip out the parts you don't need, or customize the kernel for performance.

You most certainly can do both. It costs money, of course, but remember that we're not talking about trivial tweaks like compiling the kernel for your particular processor family. We're talking about hiring a team of programmers to extensively customize the kernel so it runs your specific application and nothing else. That costs a bucket of money, and compared to that the cost of a Windows source code license is not going to be a whole lot.

I still feel that Linux would be a good bit cheaper, but we're talking big bucks both ways. And it's also worth mentioning that Microsoft's licensing model for "corner cases" like this is extremely flexible: they may give the source away at a significant discount just for the publicity. They've done it plenty of times before. Some of those 2500 servers at work run a custom-built NT kernel and we sure aren't a huge international company.

Re:Windows clusters don't make sense

[mtenhagen]

Windows does still have a GUI loaded which uses up memory and cpu cycles.

Re:Windows clusters don't make sense

[Bulln-Bulln]

Huh? Mac OS X is also very GUI-oriented, but that doesn't make it bad for clusters. I've only read positive feedbacks about Apple's Xgrid. http://www.apple.com/acg/xgrid/
So that's not really a reason why a Windows Cluster won't make sense.
Licensing costs are also not the biggest concern from big corporations.

Re:Windows clusters don't make sense

[tesmako]

This has to be some sort of world record in bullshitting, the NT kernel is easily as modular and flexible as the Linux kernel. Microsoft can easily strip and optimize the kernel (the xbox is an excellent already-existing example, great use of the NT kernel).

A number of years ago it was possible to actually listen to technical arguments on slashdot, but it seems that all technical considerations has been deemed less important than slamming Microsoft at every turn.

NT will work great in such a setting, if anything it is the historic WIN32 GUI stuff on top that is holding it back.

Re:Windows clusters don't make sense

[mtenhagen]

Microsoft indeed can optimize the kernel. But I want to change to kernel to MY application.

And no I wont hire microsoft to do that, I want to show my optimizations to several people who can critize and improve it. I want the source and the ability to improve it!

Re:Windows clusters don't make sense

[tesmako]

True enough, can't just invent it oneself. But the point here is; Microsoft is in a good technical position to make a clustered computation system. It is also most likely so that for the average user the extra configurability of Linux is fairly useless (how many people have significantly changed the Linux kernel in some way? The basic compilation settings are mostly stuff that are configurable in NT by other means anyway).

I am not saying that Linux is a worse deal in any way of course, it just does not have any real technical advantage for this specific use. On the non-technical side it is still cheaper and the added control over it gives a nice safety (also the clustering software is already here, maturity will most likely count for something for quite a few years). Claiming that the NT kernel is unmodular and an unsound base for a clustering system is just wrong though.

Re:Windows clusters don't make sense

[jfengel]

Does that require Windows Server? I was bred on Unixes of various sorts but spend a lot of my time on Windows. There are a number of operations I don't know how to do without a mouse, even on my own machine, much less on a remote machine.

Even if I had a SSH/telnet-driven command prompt, I don't think I could kill a process on a remote machine, for example; I can do it only via the GUI. Is it just because I have a lot to learn, or is it a feature I don't have?

Re:Windows clusters don't make sense

[Kent Recal]

... if there was some secure, reliable method of ...

Do you know any single piece of software made by MS that you would
consider secure or reliable?

By your definition of workstation-clustering Windows already has the feature anyways.
We're reading about its RPC capabilities twice a week aren't we?
Just remotely inject your trojan of choice (sth like back orifice?) across the office and do whatever it is you want to do. Cluster computing needs application-support anyways (unless they come up with a MOSIX which would make me laugh hard) so doing it that way is probably even easier and more secure (BO uses encryption) than whatever crap patch MS comes up with.

Re:Windows clusters don't make sense

[peragrin]

yet each of those servers still run the GUI. Just because you can't see it doesn't mean it's not in the memory.

At least with Linux you get two options. a low memory command shell that shuts down when you log out, or X which only loads the application on the local processor, using the remote machine for actual display. And when your done it turns off, restoring memory to the system.

Windows GUI never shuts off. It's always there.

Re:Windows clusters don't make sense

[Anonymous Coward]

Their "Real Application Performance" section starts off talking about how many gigaflops they get per cpu on a 4-way SMP Itanium box. This is distributed? The rest of the applications covered are at least distributed, with respectable scalability for 12 p2-300's networked with Myrinet. (Odd that elsewhere, they talk about what kind of hardware to use, and say that fast ethernet scales to 16 nodes depending on the application, I guess whatever they ran to get that graph wasn't one of those applications. There also appears to be nothing on the site about channel bonding for scaling networking capability up)

It's interesting to see there are companies other than microsoft providing distributed computing services (these people seem to favor Verari Systems' MPI/Pro, which seems to have a pretty nice range of features and runs on several platforms). I wonder if Microsoft's built-in offering will make them squeal about antitrust lawsuits.

Re:Windows clusters don't make sense

[RotateLeftByte]

IMHO, Microsoft won't sell a stripped down system as it goes totally against all their marketing spin that says "IE/WMP/etc it totally part of the O/S and we can't strip it out" Besides the deluge of lawsuits they would get lumbered with, they would risk becoming the next SCO (Derided by 90%+ of the IT Community) and suffer a total loss of whatever credibility they have left.

Re:Windows clusters don't make sense

[Krisbee]

It's not because Windows is GUI oriented that it isn't suitable.
It's beacuse it is still single-user-oriented. If you cannot be root (ahem, Administrator) you cannot do much with it.
That's the difference between Windows and MacOS X.

Re:Windows clusters don't make sense

[heffrey]

>>Which computer is running that GUI?

Who needs to run a GUI? Just use a different shell. Anyway, it's not as if running a GUI drains the machines resources anyway. It's only when the user is interacting that resources are used.

>>REmote desktop basically takes snapshots of the desktop that is running in memory

Who said Remote Desktop? RD is like VNC which runs on *NIX. Does that make VNC useless? No. RD would be one way to do remote access. Another would be telnet. Or is telnet so complicated that it doesn't work on Windows? No, I didn't think so.

Grid computing is not actually about GUIs. The ultimate ancestor said: " Windows is an overly-bloated OS which is very GUI-oriented". But Linux running a KDE or GNOME desktop is pretty similar. Why does that argument not sink Linux? Because it's not an argument. As for bloat, isn't X a little on the fat side?

Yawn.

Re:Windows clusters don't make sense

[azaris]

Even if I had a SSH/telnet-driven command prompt, I don't think I could kill a process on a remote machine, for example; I can do it only via the GUI. Is it just because I have a lot to learn, or is it a feature I don't have?

rkill, but I think it's an installable service that only comes with Resource Kit.

Re:Windows clusters don't make sense

[FooBarWidget]

That's as valid as "Linux is ready for the desktop". Until you Slashdotters accept that Linux is ready for the desktop and stop trolling, Windows will never be ready for the cluster!

Re:Windows clusters don't make sense

[peragrin]

>>But Linux running a KDE or GNOME desktop is pretty similar.

of course it is but Linux doesn't need a gui. How do you tweak your registry settings without a gui? updates, okay they can be done, but how do you fix the little things when you have to run the GUI to do so.

yes remote desktop is like VNC, but then again you don't need VNC to run linux.

I guess the simple solution is show me a telnet(ssh) server that gives you FULL control over the entire OS from the command line. Not just the ability to launch a remote desktop system.

And windows GUI is built into the kernel at least parts of it. tech wise the NT kernel isn't bad, it's a lot of extra stuff that MSFT has added in.

Also note why is windows using telnet? unless it's a totally closed network(which it should be anyway) it should be running ssh.

Re:Windows clusters don't make sense

[Anonymous Coward]

I think the point is that the GUI is running on those remote systems, whether you use it or not. And those GUI processes eat up memory and processor resources that would be better used to run whatever tasks are assigned to the cluster.

My Linux cluster runs an OS that is trimmed to something less than 32M on each node, leaving the rest of 256M RAM space for processes running on the node. Windows simply cannot match that.

In my cluster, each node is diskless. They all load a copy of the OS upon power on and run self-contained in RAM. Disk activity is mapped to a central server. Aside from the difficulties of booting Windows diskless (is it even possible?), the insane amount of disk activity that Windows requires would tie up communication resources that, again, would be better used by the processes that run on each node.

Economically, Windows makes no sense in a cluster environment either. Microsoft's insane licensing arrangements would cost me almost as much as the cluster itself cost just to run a copy of Windows on each node.

Now, let me ask you: Do you know anything about Windows?

Re:Windows clusters don't make sense

[diegocgteleline.es]

You don't need to have a GUI to run NT. It doesn't sounds like something hard to implement - just code a "dummy driver" which does nothing.

BTW, windows embedded allows customers to "customize" what parts of the kernel they want. They can do the same with a "cluster oriented" windows version

And remember, even if windows sucks for cluster it don't means they won't have success. Windows 9x was crappy base to build a OS on it, despite of that everyone bought windows 95.

Re:Windows clusters don't make sense

[Bulln-Bulln]

An idling GUI takes almost no CPU performance and, if well configured, the memory consumtion of a GUI is not so big either.

Re:Windows clusters don't make sense

[heffrey]

ssh runs fine on Windows. You can edit registry from command line no probs.

Re:Windows clusters don't make sense

[heffrey]

And you can turn off Windows GUI.

Re:Windows clusters don't make sense

[heffrey]

Well, AC, here goes.

Why would a Windows distributed node need to run a GUI? The Windows kernel can function quite happily without a GUI.

Windows can be trimmed down. Remember that this is a future project so MS can do what they want. They do have the Windows source after all. No reason why they can't get a pared down OS.

Diskless booting of Windows is possible today.

Economically we don't know what the licensing for a distributed version of Windows would be since MS hasn't decided that yet - this product does not yet exist after all!

I'm not arguing that distributed Windows clusters will be a great success. All I'm saying is that this GUI stuff is not relevant. If a GUI eats resources then the product would have a console only mode.

MS has got a decent kernel, lots of experienced, clever developers and stacks of cash - it ought to be able to make this work if it so desired.

Re:Windows clusters don't make sense

[paganizer]

Solitaire? very secure and reliable.
Pinball is pretty good, also.
I like the VPN client/server system built into NT-Win2k. it's very reliable, and the security problems it has are fixable with a little work.
Exchange 5.5 mail server is robust and pretty secure.

Re:Windows clusters don't make sense

[skraps]

You don't know much about virtual memory, do you?

Re:Windows clusters don't make sense

[Bishop]

I dislike Windows, but most of what you wrote is wrong.

The Windows GUI can be turned off, along with many of the other services that you won't need in a cluster. It is not even that hard. The MS knowledge base is a mess, but the information is there. There are many performance tweaks for the NT kernel that don't require a recompile. It should be noted that most Linux clusters use unmodified, or lightly modified kernels. Most admins feel that the slight performance gain (if any) is not worth the maintanance. While a Linux license could cost $0 most clusters pay more. The cluster owners want the maintanance that comes with a commercial Linux distro. The Windows licenses are actually not that expensive, and MS is also more then willing to negotiate a better price. Linux would probably cost less, but not that much less.

Finally there are some nice tools for Windows that allow you to manage a large cluster. Tools that are more sophisticated then ssh or distributed ssh. Tools similar to the ones that RedHat and Suse are busilly writing because the equivalent Linux tools are pretty basic.

There are many reasons why Linux is perfect for clusters. But none of the reasons you list are valid.

Re:Windows clusters don't make sense

[Bulln-Bulln]

It's beacuse it is still single-user-oriented. If you cannot be root (ahem, Administrator) you cannot do much with it.

Of course you can do much with it. Problems with non Admin accounts result from bad Win9x programming habbits. If the developers take care and write their apps in a correct way.

I don't think that the whole system needs to be rewritten from scratch to support clustering. Maybe 5% of Windows (Just a guess).

I don't say that Windows will be the best clustering OS ever, but clustering on Windows is possible.

Re:Windows clusters don't make sense

[fymidos]

>Diskless booting of Windows is possible today.

can you explain that? how can windows boot without disk (and which version of windows are you reffering to ? i imagined that diskless booting of XP must be impossible)

Re:Windows clusters don't make sense

[fymidos]

still his point is that it doesn't make sense, although they *can* be used for clustering, they are simply not the right tool for the job. This is not simply my opinion, the market has spoken on this:
They cost money, they provide no advantages over linux or bsd, and they would propably need much more human hours for their administration.

besides, as i understand the article, this is mostly a development thingie, and some sort of central application management service. And with a possible* release date somewhere in the end of the decade it just doesn't seem important. I don't understand why this story was even posted ...

*possible as in "maybe it will be released" not as in "maybe in the end of the decade"

Re:Windows clusters don't make sense

[Kent Recal]

A VPN-system that comes with security problems "fixable with a little work"?
Exchange robust and "pretty" secure?

You deserve +3 funny ;-)

Re:Windows clusters don't make sense

[heffrey]

Try Google. I searched for windows xp diskless booting (http://www.google.co.uk/search?query=windows+xp+d iskless+booting&hl=en) which takes you to http://msdn.microsoft.com/library/default.asp?url= /library/en-us/dnxpesp1/html/tchDeployingWindowsXP EmbeddedRemoteBoot.asp amongst many other links.

Re:Windows clusters don't make sense

[Junta]

I seriously doubt Windows pages out much of the resources regarding the GUI. That is part of the reason why X can get really sluggish under memory intensive operations (users of X occasionally can really tell when something is getting read into memory from swap), and Windows GUI tends to stay more responsve....

That said, honestly the resources the GUI takes are largely a red herring, the resources the GUI requires if truly left idle are not very significant in the scheme of modern systems' resources. And besides that, I have seen at least a couple of Windows systems where the GUI didn't actually start locally (it is strange, the boot progress splash screen stops, as if the system locked up, but that is when the system is up and ready).

Of course, I strongly prefer linux for clustering applications. The whole system is very manageable via nothing more than a serial console or ssh. It lends itself better for scripting, and the debatable advantages that Windows may possess are a moot point in the environment. If both your OS and BIOS work great with a serial console, Serial Over LAN or, worst case, serial terminal servers are much cheaper than KVM solutions for large numbers of systems, particularly network accessible KVMs.

Re:Windows clusters don't make sense

[The_Spud]

A friend worked on a project which used windows XP embedded and booted off flash memory set to read only. He managed to strip out unnecessary bits of xp and got the install size down to around 200MB if I remember correctly. It was for a museum display system. Shown here

http://www.hunterian.gla.ac.uk/collections/museum/ scientific/index.shtml/
If you have the XP embedded development kit you can get XP to install on all sorts of media.

Re:Windows clusters don't make sense

[fymidos]

The parent was talking about diskless boot, thats when you have a machine with a nic and memory, -no hdd, no floppy, no flash-, and you download *everything* from the server.

My question was whether that was possible with some version of windows. apparently it is for winxp embeded sp1 as i see in his link:

>.. Windows XP Embedded with Service Pack 1 >introduced the Remote Boot feature ...

Re:Windows clusters don't make sense

[fymidos]

i checked, and apparently there are some third-party solutions for windows diskless booting.
That microsoft link though, only talks about booting winxp embedded, for diskless gadgets, or for firing up a terminal server client (hardly usefull for a cluster where you need the clients' processor to do the math).

Re:Windows clusters don't make sense

[Bas_Wijnen]

So you're saying that if we just don't accept that GNU/Linux is ready for the desktop, Microsoft will stay off clustering? Great, I just changed my mind and no longer think GNU/Linux is ready for the desktop then. :-) But feel free to use it anyway. ;-)

Re:Windows clusters don't make sense

[value_added]

"We have some 2500 Windows servers where I work. None of them have monitors, keyboards, or mice. If we need a KVM it's typically to get into the BIOS, not the operating system."

Sounds nice, but what you're really saying is that a Windows server will load and run without certain hardware peripherals attached. Ignoring the event log errors issue, I don't see this as meaningful.

1. You cannot administer a Windows server (or a desktop) to any meaninful extent on the command line. You can add all the resource kit utilities, sysinternals utilities, whatever other little programs the Windows folks routinely get excited about to the embarrassingly small collection of native "DOS" commands, and then throw in all the vb scripts you can write, and you still end up with a toolkit the size of which would maker any Viagra marketer drool.

2. That Longhorn may introduce some shiny new approach to a terminal window (non-text-based from what I've heard) is saying that it won't be as crippled as cmd.exe, but not much else. I doubt anyone is holding their breath. And I doubt anyone believes that Microsoft will get it right.

3. Changing the Windows shell (to cmd.exe, for example, as was previously suggested) is little more than a distraction. You swap out a visual desktop that displays the start menu, taskbar, system tray, etc., for a black background. Everything else, with a few quirks here and there is alive and well as before.

4. That you use KVMs mostly to "get into the BIOS" also says little. It would have been more informative to say that your company manages its 2500 servers with terminal services or something similar and saved money by not buying a monitor for each sever.

Re:Windows clusters don't make sense

[Foolhardy]

top - pslist
ptree - pslist
w - psloggedon
ls -al - dir /adhs, fsutil (both standard)
finger - finger (standard)
unzip - expand (standard, for CABs), cygwin unzip, rar
mount - (automatic), fsutil, linkd (from resource kit)
make - make (comes with SDK)
grep - find (standard)
piping with | > < are the same
perl
cygwin for other UNIX processing utils.

Re:Windows clusters don't make sense

[ClosedSource]

Well, I think your definition of "diskless" is too restrictive. It's OK to have RAM but not flash? I'll bet there aren't any Network Computers without flash or ROM.

If you meant a "Network Computer" you should have said that instead of "diskless".

Re:Windows clusters don't make sense

[ReeprFlame]

Microsoft's incredibly expensive and anal licensing makes a Windows cluster not worth the effort or money. I mean, Linux's licensing cost is 0, and 0 scales infinitely Good quote. Goes along with the bullshit info they sent me on Windows Server 2003 and why it is better and *cheaper* than Linux. 0 to any extent is still zero with alot of public support and free updates and patches [did I mention patches are made ASAP]. Linux always dominates. HEll, when I ran my server on WinServer2003, it crashed almost everyday without intervention. Linux has been stable until I have had a power outage!

Re:Windows clusters don't make sense

[ReeprFlame]

That's called power down. Or back in Windows 98 "Restart in MS-DOS Mode". Ah, DOS, the anti-BSOD OS...

Re:Windows clusters don't make sense

[ReeprFlame]

I've used the IIS 6.0 web application bundled with Windows Servers [2003 that is]. I have to admin that it is rather nice. Similar to webmin, but still a bit better in some respects...

Re:Windows clusters don't make sense

[gothamboy]

The real question is why a non "huge international company" needs 2500 servers for running its business! Seems to me if they ported their ap to Linux, they could get away with only 2450 servers!

Re:Windows clusters don't make sense

[priich]

taskkill.exe, tasklist.exe and sc.exe for simple operations. For more power, wmic.exe is the tool of choice.

Re:Windows clusters don't make sense

[SunFan]

If we need a KVM it's typically to get into the BIOS....

Ah, the pleasures of OpenBoot PROM and a serial cable...

Re:Windows clusters don't make sense

[4of12]

Windows isn't much use there. Nor can you strip out the parts you don't need, or customize the kernel for performance.

I'm positive that using a stripped down version of XP embedded that MS could release something that could do this job well. They could price it so reasonably that people could put it on as many nodes as they wanted, or run as many instances on virtual machines as they wanted. MS could even open up the APIs so completely that people could do more of the customization that they do now on Linux.

But even though they are technically capable of doing this MS won't do this because it will collide with their business objectives elsewhere.

I think it's all a PR thing - kind of like Ford sponsoring autoracing, with the idea that people buying the Escorts will think they're getting part of a Formula One racer.

could be good

[Cheeze]

i always wondered why there's not an easy way to utilize all of the computers in a network to perform a task. Most of the computers on corporate networks are windows machines, and most of those are sitting idle 99% of the time. If there was a way to harness that power for something useful, like an oracle database, web hosting, mail hosting, etc, the whole network would not be bottlenecked by one overloaded server. Mosix kinda solves that problem, but on the linux-side only.

If someone wanted to make millions of dollars, build something like that for windows and charge minimally for it. Better do it before Microsoft does.

Re:could be good

[basvdlei]

I'm not an expert but using employees workstations to power your companies database sounds like an security nightmare to me.

But in there are indeed cases where it would not be such a bad idea.

Re:could be good

[Cheeze]

If they are using the network, they already have a security model in place. There shouldn't be any additional security risk offloading cpu/storage to client workstations. Even if you offload only to other servers, the advantage of being able to add an additional server to increase cpu horsepower or storage space is invaluable. Most large sites already have this in place, they just had to build it in-house (like google, amazon, ebay).

I'm sure Microsoft will build something like that, and it will totally suck. Then someone else will come to market and their application will totally dominate. Then Microsoft will buy them, and the application will end up being like Virtual PC or something.

Re:could be good

[LiquidCoooled]

Spyware and trojans have been doing exactly this for years now.

Re:could be good

[mjh49746]

Why do it before M$ when all M$ has to do is embrace, extend, and extinguish like usual? Doesn't matter who does it first, because in the end, only M$ will end up doing it, and they'll have a bunch of patents to boot just to stymie open source and to make sure no one else will be able to do it. We all know how the USPTO works, and they won't really look for prior art, either.

Re:could be good

[AaronGTurner]

"i always wondered why there's not an easy way to utilize all of the computers in a network to perform a task. Most of the computers on corporate networks are windows machines, and most of those are sitting idle 99% of the time."

Specifically for Oracle there is Oracle 10g.

For various other classes of computation there are the following (plus others) on windows (and some are cross platform):

* Condor
* Entropia
* United Devices
* BOINC
* IBM community grid
* Vita Nuova's Inferno
* Sun Grid Engine (coming soon)

I'm not aware of things that specifically do load balancing on Windows as opposed to CPU scavenging or batch job submission. My work on Grid systems hasn't addressed this for Windows. I can see ways that some of the tools can be used to achieve load levelling and scavenging combined.

Re:could be good

[Just Some Guy]

I'd like to see distributed applications on a much wider scale. Right now my office is fully of extremely-underutilized 2.4GHz P4s. If it were possible to share the office's processing and storage conveniently, we could get by with an office of P-II 333s that were running at 50% capacity instead.

Most of the machines are doing word processing, email, and other light-load activities. Very slow individual machines aren't useful because sometimes users want to run applications that require large amounts of CPU for a very brief period of time. However, if Jane Secretary could borrow the collective CPUs of 30 slow integrated machines for a couple of seconds, her application would run as fast (or faster) than if she had a high-end system next to her desk. As an added bonus, you could upgrade the performance of every machine in the office by adding a couple of fast machines into the cluster.

I realized that I just described the benefits of client-server computing, but maybe it's time for a new common paradigm (OMG, I can't believe I just used that without intentional irony): client-to-client computing.

Re:could be good

[lachlan76]

If they are using the network, they already have a security model in place. There shouldn't be any additional security risk offloading cpu/storage to client workstations

But this would mean that clients have access to data that they shouldn't be allowed to see. On a neteork with client/server, the client doesn't get to see that data.

And would it be secure if everyone had physical access to a server with the current system, anyway? No.

Who wrote the summary?

[John Harrison]

As of now it is unclear, but Microsoft probably will bring this situation to life in the near future since it does hold alot of power for them over other platforms.

Does this make any sense? The rest of the summary is equally nonsensical.

Re:Who wrote the summary?

[Scott7477]

Here is tha actual article; note that MS doesn't plan to have this ready to release until "near the end of the decade."

A Peek Under Microsoft's Secret 'Bigtop'
By Mary Jo Foley, Microsoft Watch
December 29, 2004

Microsoft officials have said little about the company's intentions in the grid-computing space. But that doesn't mean Microsoft is ignoring the evolving arena of grid/distributed computing.

Microsoft is working on a skunk-works project that is code-named Bigtop, which is designed to allow developers to create a set of loosely coupled, distributed operating-systems components in a relatively rapid way, according to sources close to the company, who requested anonymity.

Rather than attempting to tightly couple a few high-performance systems together, Microsoft is looking at the consequences of loosely coupling a larger number of moderately powerful computers to achieve a similar result.

Bigtop's first commercial manifestation will likely be as some kind of large-scale project, most likely a distributed grid-computing operating system, the sources added.

Bigtop is one of Microsoft's incubation projects. It falls under the domain of Craig Mundie, the Microsoft senior vice president and chief technical officer in charge of advanced strategies and policy, sources said.

Bigtop consists of three components, all written in C#, according to developers who said they were briefed by Microsoft. These are:

Highwire: Highwire is a technology designed to automate the development of highly parallel applications that distribute work over distributed resources, the aforementioned sources said. Highwire is a programming language/model that will aim to make the testing and compiling of such parallel programs much simpler and more reliable.

Bigparts: Bigparts is code designed to turn inexpensive PC devices into special-purpose servers, according to the sources. Bigparts will enable real-time, device-specific software to be moved off a PC, and instead be managed centrally via some Web services-like model.

Bigwin: According to sources close to Microsoft, Bigwin sounds like the ultimate manifestation of Microsoft's "software as a service" mantra. In a Bigwin world, applications are just collections of OS services that adhere to certain "behavioral contracts." These OS services can be provided directly by the core OS or even obtained from libraries outside of the core OS.

Sources said Microsoft will likely make some sort of preview version of the Bigtop code available to the company's software-development partners by 2006. If and when the final version debuts, it won't be much before the end of the decade, sources added.

It's not clear whether the Bigtop components will run on top of Windows when they are completed. But sources say that is what they are expecting at this point. End of Article

I like their use of a circus term as a name for this project. It gives the impression of a bunch of clowns running around into each other and falling down. Kind of like MS systems on the web now.

Re:Who wrote the summary?

[AaronGTurner]

"Bigtop's first commercial manifestation will likely be as some kind of large-scale project, most likely a distributed grid-computing operating system, the sources added."

This has already been done (Inferno network OS from Vita Nuova, which is very nice to work with and can also be run hosted on other OSes)

BigWin sounds like wrapping of traditionally OS components in terms of Web Services and hooks to use these. Microsoft has strongly embraced Web Services. This is an entirely obvious development for anyone skilled in the art (Grid Computing).

Sun GridEngine

[Anonymous Coward]

Gridengine just added Windows support:

- Windows XP and 2000 (December 2004 availability)

http://www.sun.com/software/gridware/

Gridengine's source can be downloaded from:

http://gridengine.sunsource.net/

Re:Sun GridEngine

[marknmel]

Ya, I saw that on Sun's web site. I have my doubts though.

I have been fighting about this with SMI since August/04. That date keeps floating. Last I heard was April/05 for FCS, and perhaps technical beta sometime in Q1/05.
Bah!

That will be just like Slowaris 10, the launch was in early November/04, but we won't see it 'till March-0April/05. More vaporware from Sun!

They already dominate grid computing!

[Spackler]

Just plug an unpatched XP box into the internet. It will be part of the worlds largest grid computer in less than 2 minutes.

It will also hum the tune Zombie Rock!

Re:They already dominate grid computing!

[ElQueso]

Or, if you want to be on the safe side, plug a patched computer, but it will take longer to join the grid, something like 10 minutes

Awesome

[knightrdr]

This is just what we need for a grid... an OS that will consume 10% of system resources just playing solitaire.

Re:Awesome

[ngyahloon]

an OS that even has a process just for idling (System Idle Process)

Re:Awesome

[tabrisnet]

Actually, pretty much all do, but it's really just for accounting purposes.
Tho it should be noted, that at least in linux, the 'idle process' can actually be doing work. Servicing hardware interrupts. However it should also be noted that _any_ process can get charged for hard-irq servicing. Of course, the cost of the hard-irq service is kept to a minimum, and as much as possible is pushed into a deffered 'soft-irq'.

This is done mostly b/c at hard-irq time, it is largely impossible to know beforehand which process the work really belongs to, but in theory a CPU hog should be causing most of the IRQ work as well. Obviously this is not always the case, but it is the fairest that is possible given the architecture, as well as getting hardirqs serviced as quickly as possible.

Already done

[Progman3K]

There are millions of Windows machines out there participating in a distributed SPAM relaying network.

I imagine if Microsoft 'enahances' Windows to do this even easier, it'll make it even easier for spammers to write the next-generation spamming-joe-jobbing apps.

Kudos, Microsoft!

Leaking as a business tactic

[Punctuated_Equilibri]

Seems to me this is a deliberate leak to create uncertainty in customer's minds and block any adoption of *nix for grid computing.

IT decision makers can be persuaded to wait for Microsoft's solution rather than 'taking a chance' with another OS.

Re: Leaking as a business tactic

[Black Parrot]

> Seems to me this is a deliberate leak to create uncertainty in customer's minds and block any adoption of *nix for grid computing.

That, or they're priming consumers to accept the idea that it will take a whole rack of computers to run the next version of Windows.

Re: Leaking as a business tactic

[gstoddart]

That, or they're priming consumers to accept the idea that it will take a whole rack of computers to run the next version of Windows.

Wait until you see the recommended (as opposed to minimum) equipment. ;-)

Besides, it seems flaky enough when it's just one machine. I can just see a group of them collectively borking each other at random times for no discernable reason. ...

"You're other computer has done something, reboot now?" Doh!

hardware is the cost

[mtenhagen]

When you want to do large computations the biggest cost is the hardware. So you want to make optimal use of your hardware by using software optimized for that hardware. Rewriting networkcard software can give you improvements of 10-20% for your specific application.

On linux you can remove interrupts from the kernel if your app only needs polling. Stuff like that will never be possible with a closed source solution.

Lots of ppl stop using solaris cause of this.

Re:hardware is the cost

[ClosedSource]

"On linux you can remove interrupts from the kernel if your app only needs polling. Stuff like that will never be possible with a closed source solution."

Actually, closed source real-time OS's have allowed that kind of stuff for years.

Re:hardware is the cost

[SunFan]

Lots of ppl stop using solaris cause of this.

Solaris will be Open Source in a month or so. Enjoy.

confusing parallel and distributed computing

[goombah99]

the article poster seems to confuse parallel processing on a single machine with distributed computing. The difference is that each machine is running it's own OS and not sharing physical memory in distributed computing.

distributed computing happens at the application layer. Thus if you can run something like an MPI library on windows you have the basis for efficient distributed computing. All you need is a scheduler and launcher to be able to launch distributed launch an application across the net. But virtually all of these are daemons not strictly part of the OS. So that level of system independent abstraction exists already so this should not be too difficult.

Re:confusing parallel and distributed computing

[pottymouth]

You CAN run MPI (and PVM) on Windows. You can even run a processes that share cycles between a Windows and a Linux machine. MPI and PVM are both capable of working on a heterogenius network (though MPI didn't start out that way).

Re:confusing parallel and distributed computing

[Anonymous Coward]

Yeah, I thought that was odd. There already are several distributed apps that run on Windows. People just usually don't do that because they don't want to pay for licenses for each node.

Re:confusing parallel and distributed computing

[abradsn]

Right,
I'm not sure why people don't think you can do distributed stuff on windows. I do it all the time in my job and it works fine.
Windows has a lot to provide in this area, and it just gets better and better. Granted it could have been better a long time ago. (Cough, Cough, DCOM, gag)

No ETs yet...

[ScentCone]

...but my SETI@home screen saver is one of the most stable apps on my XP machine. It certainly doesn't qualify as "grid" computing, but it feels awfully big some days.

Of note: I've got some Win2K web servers running in a native WLBS load balanced rig, and those machines have been doing swell for four years now. They talk to a cluster of SQL servers, but that clustering really doesn't count... it's more like hot fail-over. The native load balancing of the web servers, though, has been pretty tight and has scaled very easily, at least within my mid-market universe.

I know, I'm just asking for it with this post. Just wanted folks to know that it's possible to push a couple $million of holiday e-commerce through some pretty cheap white boxes running MS's stuff. And yes, my cheap admin help is glad there's a GUI for some of the chores they don't do every day. All right, flame me now. But you have to do it from a command prompt.

Re:No ETs yet...

[Libor Vanek]

Hi, could you post (maybe to email - this is getting a bit off-topis) some more info? What FS do you use? (OpenGFS from Sistina/RedHat?) How kind of KDE applications and how do you balance them accros nodes? Thanks.

Re:No ETs yet...

[codepunk]

Actually the raw desktops are kde all of the custom
apps we run are qt based programs built with borland kylix.

Compaq Quad Boxes
CentOS with GFS and Mosix Kernel
Hitachi SAN shared fiber channel scsi

To get a great and simple introduction to the architecture just grab a copy of Kluster Knoppix
and boot about 10 nodes and play with it. Our environment is slightly more complicated but it
is esentially the same thing.

Re:No ETs yet...

[Libor Vanek]

Yeah, but how do you balance the KDE applications - do you only load-balance (start & run application on "random" node) or you can migrate and/or duplicate (for fail-over) processes?

Re:No ETs yet...

[Bert64]

But for this you need to use ms's "server" os's which push up the price of the cheap white boxes quite considerably..
In essence, you need to have a smaller number of faster machines (slower machines may not even run these os's), with a linux/bsd cluster you could just add cheap/old machines at will.. I've done that before, i have a cluster of 4 machines running 24/7 with 16 old p200's on standby that get booted up during busy periods.

Re:No ETs yet...

[codepunk]

The mosix kernel will migrate the process to the least loaded node. You can manually migrate them if you wish using the mosix admin tools.

Longhorn, Grid, RFID, SmartCards...

[MosesJones]

Microsoft are wonderful. Trying to sell things they don't have in order to make it look as if they are ahead of the pack.

So if its announced in 2004/5 it will be "scheduled" for launch in 2007, but actually arrive in 2009.

Re:Longhorn, Grid, RFID, SmartCards...

[bob beta]

Yet again, another innovation that Microsoft 'stole' from others. Larry Ellison's army of marketers pioneered in the area of vaporware. Microsoft just followed in their footsteps. They aren't even as good at being slimy marketers as Oracle.

Re:Longhorn, Grid, RFID, SmartCards...

[tomhudson]

The article said they plan to have it out by the end of the decade ... they didn't say WHICH decade, so we're probably looking at 2019, or maybe 2038.

It's (the announcement) a trial balloon, along the usual Microsoft marketing lines (throw enough shit, some of it will stick). They did the same thing for years with the original versions of Windows, to try to keep the market from adopting competitors.

Of course, they're too late - the free OSes have them beat already, and by the time Microsoft comes out with something, hardware will be so cheap you won't need it - just make your own *nix supercomputing cluster with the money you save on licensing.

Insightful?

[glrotate]

Networking was an add on to NT/2000/XP/2003? Hardly.

How about a /. New Years resolution? Cut the ani-microsoft hyperbole in half?

Windows has been clustering for years

[skinfitz]

Q. What do you call a cluster of Windows machines?

A. A botnet.

Re:Uh, right

[codepunk]

And If your friends where smart they would use linux to do CFD. We switched from windows to linux in our CFD lab and cut solution computation time nearly in half. On top of that we no longer have system crashes that ruin the results.

cutting possible

[spectrokid]

they do make an "embedded" version where you can cut crap away. Not that I think this is going to fly though... I like the poster: Would it be an addition to home users computers as well as the server versions of Windows? Considering all non-server versions only accept one telnet client at a time, keep on dreaming buddy. This is probably gonna get marketed as a distributed server thingy for large web or exchange clusters.

hmm interesting

[fmobus]

wow imagine a beowulf cluster of these... oh wait...

Not entirely accurate

[Sardak]

Considering that the Windows platform has never had the ability to parallel compute in the past, it leaves great potential to the company's operating system development.

This isn't entirely accurate. Server 2003 supports clustering, and if I remember correctly it even has a toolkit that allows you to add XP Pro systems to a cluster started by the Server 2003 machine. A friend of mine created a small cluster as part of one of his courses last year.

Re:Not entirely accurate

[Tony Hoyle]

That is *not* a cluster. It's a load balancing server.

MS like to call it a cluster because it makes them sound 'good', but really it's crap.

MS clustering? Its a Joke!

[Savage650]

Keep in mind though that Windows clusters are existing.

Looking at the MSFT definition or clustering, they describe two kinds of clusters:

• network load balancing clusters ("[the type ..] that distributes and load balances network connections among servers, providing high availability and scalability for stateless TCP/IP applications and services.").
  Note the explicit restriction to "stateless".

• server clusters ("[the type..] that the Cluster service implements. Server clusters are characterized by high availability.)
  Note they mention availability but not performance.

ObJoke: MSFT renamed "Wolfpack" to "Server Cluster API", probably because they were sick of people describing it as "two dogs fucking" (As in: two beasts stuck together, pulling in opposite directions and howling in pain).

Exactly What We Need

[Bruha]

A giant Windows machine cluster that gets a virus.

Nothing to worry about here...

[BrokenHalo]

In a world of distributed computing, who is going to wear the cost of multiple and/or site licences when they have already forked out for development of the distributed apps in the first place?

Seems to me that the open source platforms are well and truly set to crucify MS in this market. Why pay for a platform when you don't have to?

Re:Nothing to worry about here...

[ReeprFlame]

Yes those licences will get you every time. MS will probably even make you pay extra for each slave computer especially beacuse its part of a cluster. That's why I will pirate my copy!

Last time I checked...

[Ancient_Hacker]

Last time I checked, Windows in all its multifarious versions has no way to run a program in a sandbox, such that this program is incapable of DOS'ng the PC by opening tons of windows, file handles, memory blocks, processes, etc.... If the system isnt designed fromt he ground up to be compartmentalized, stable, and secure, IMHO there's little change of grafting all these qualities on a decade down the road.

Re:Last time I checked...

[Ancient_Hacker]

Okay, there may be limits, but in the XP Pro distro, they're not too sanely set. For example, the batch file go.bat: ---------------- :loop start go.bat goto :loop ---------------- Give it about 15 seconds and it has 184 copies running, true not all got to open console windows. But the system is pretty well hung, ctrl-alt-del can't get enough priority to bring up its window, the GUI is inoperable. Had to turn off the power to stop all the commotion.

Re:Last time I checked...

[spectecjr]

Last time I checked, Windows in all its multifarious versions has no way to run a program in a sandbox, such that this program is incapable of DOS'ng the PC by opening tons of windows, file handles, memory blocks, processes, etc....

Perhaps you should check again, and learn what Job Control Objects are used for.

Summary of posts

[Anonymous Coward]

Summary of every post in this topic:

This is bad. M$ is evil evil. *Cough* . Bloated, FUD, GUI, copied MAC, FUD, [nonsensical, nonsensical] bloated, *Cough*, I'm waisting my life ^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H. I can't believe people are so stupid to belive such M$$ lame FUD, propoganda [ nonsensical... ] Blue screen, Blue Screen!. Linux good. Why are M$$$ so stupid? Ha Ha, I'm so much smarter. *Cough* Blue Screen! this is like Clippy! [nonsensical, nonsensical], really crap. Mac good. Bad idea, unstable. Blue Screen! Open Source, Open Source! [ nonsensical... ]. M$ Bob. Zombie. Blue Screen, Blue Screen! Security ^H^H^H^H^H^H *cough*. IE, ahhh! ahhh! Blue screen. Stupid.

Re:Summary of posts

[ReeprFlame]

True. I mean MS does actually make sense in at least *attempting* a porject that may result in somthing good for the community. Ok Linux has done it before, but soem apps only work well in Windows, and if you can make them moerpowerful, why the hell not do it?

Not the same thing ...

[gstoddart]

Huh? Mac OS X is also very GUI-oriented, but that doesn't make it bad for clusters. I've only read positive feedbacks about Apple's Xgrid.

In this case, Mac OS X is sitting on top of a UNIX kernel -- a modified FreeBSD. Which means all of those parts aren't GUI oriented, and you get all of the same benefits of a UNIX with all of the eye candy that Apple knows how to make work well.

Windows seems to have been built with a model that expects everything to want to be GUI based and it includes a lot of stuff geared towards that. As has been pointed out elsewhere, Windows seems to be taking networking and other stuff as add-ons without having been accounted for in the first place. Though that's probably changing somewhat over time.

In the case of OS/X, it will happily do both functions without saddling the non-GUI stuff with extra baggage.

Re:Not the same thing ...

[Bulln-Bulln]

Mac OS X is sitting on top of a UNIX kernel -- a modified FreeBSD.

Wrong. OSX' kernel is XNU - a modified version of Mach. OSX (or better Darwin) includes a lot of FreeBSD code, but it's not just a modified FreeBSD.

There's also a difference between Windows as a whole and just the NT kernel. The NT kernel isn't that bad. Most problems with Windows result from problems in the higher levels of the system - eg. IE.
Problems with higher levels of the Windows OS are not necessarily a reason against clustering on NT based Windows systems.

Re:Not the same thing ...

[gstoddart]

Wrong. OSX' kernel is XNU - a modified version of Mach. OSX (or better Darwin) includes a lot of FreeBSD code, but it's not just a modified FreeBSD.

Well, according to wikipedia it's more of a hybrid. But, yes, I concede it's not just a re-worked FreeBSD, but to a process it offers an almost identical interface.

There's also a difference between Windows as a whole and just the NT kernel. The NT kernel isn't that bad. Most problems with Windows result from problems in the higher levels of the system - eg. IE.

Well, as Stallman likes to keep pointing out, a kernel does not a full O/S make. And if you build your BSD kernel into your Mach, you deliver a more UNIX-y operating system which is exceedingly good at providing access to infrastructure things like networking and the task-swapping and the like.

Though the NT kernel might be not too bad, I'd still put my money on the behaviour of the whole platform to be better on BSD when GUI stuff isn't really involved or needed. To me the scheduler and the like in Windows always feels like it's not as adaptive as a BSD kernel -- so by the time NT becomes Windows, it doesn't feel as efficient or responsive with similar hardware.

Re:Not the same thing ...

[SunFan]

OSX' kernel is XNU...

You just enabled an entire Apple-Scientology conspiracy theory.

Close collaboration with Cornell Theory Centre

[ngyahloon]

I believe microsoft is working quite closely with the Cornell Theory Centre on this. Cornell has been using Windows for their HPC work for quite some time, concentrating mostly on financial analysis stuff. They even have a course on Windows HPC.

Microsoft is all about user choice

[bcmm]

Does that mean that when (if?) Windows Longhorn boots up for the first time, the user will be offered a list of available botnets?

That would be a major advance on the current behaviour of just selecting a botnet at random, a system that has annoyed some users.

Heh...

[BrokenHalo]

Indeed; as my fortune cookie for today reads:

A master was explaining the nature of Tao to one of his novices. "The Tao is embodied in all software -- regardless of how insignificant," said the master.

"Is Tao in a hand-held calculator?" asked the novice.

"It is," came the reply.

"Is the Tao in a video game?" continued the novice.

"It is even in a video game," said the master.

"And is the Tao in the DOS for a personal computer?"

The master coughed and shifted his position slightly. "The lesson is over for today," he said.

-- "The Tao of Programming"

Re:Uh, right

[l3v1]

Gee, so that's why my friends in the CFD group have been running flow solving codes using MPI and Windows 2000 machines!

So the ability of using a message handling library for parallel programming does indeed make Windows the tool for parallel computing. Now that's the point where I usually wake up laughing my a** off.

Not So Funny: China and its Threat

[Anonymous Coward]

This new distributed computing technology which Microsoft will incorporate into Windows must be safeguarded from China. The Chinese would use it to enhance their military capability.

In particular distributed computing has particular use in doing wargame simulations. The Pentagon has been doing its simulations on distributed systems for, at least, a decade.

DCOM, COM+ anyone?

[Otis_INF]

Windows already has distributed computing build in, with transaction support which controls cross-machine/process transactions, it's available in every windows box (2000/XP/2003). Furthermore it has object-level security settings, based on roles, integrated in for example Active Directory so you can control which user can access/run which object.

'Grid computing!!!111'... it's a buzzword. The technology is already available for many years, however not a lot of software uses it, if you look at the many many applications available.

Considering that the Windows platform has never had the ability to parallel compute in the past, it leaves great potential to the company's operating system development.
I don't know how much 'ReeprFlame' knows about windows, but it can't be a lot. :-/

Re:DCOM, COM+ anyone?

[codepunk]

Does it have a clustered file system?

Nope, shut up and go back to playing with the hot wheels you got for christmas.

Re:DCOM, COM+ anyone?

[multi io]

Windows already has distributed computing build in, with transaction support which controls cross-machine/process transactions,

Which has next to nothing to do with grid computing.

Repeat After Me

[codepunk]

All the MS fan boys on here need to repeat after me.

Windows does not have clustering!

Although they may have the capability of real clustering some day they do not have that capability today no matter how much your resident MCSE talks about his great exchange clusters etc. Windows can load balance and it can provide failover and it can run some distributed processing software but it cannot natively cluster.

Linux on the other hand has the tools available to run a true cluster, failover, load balancing and a real cluster aware file system meaning all nodes can share, distribute and balance processes. A clustered file system means that all processes even running on cluster nodes can access the same exact data.

For the linux crowd I would not exactly worry about this as MS is light years behind when it comes to this capability. We run some CFD solvers at work which where initially put on some 2K boxes and have been since migrated to linux to eliminate system crashes and improve the solution speed by nearly 50%.

We do have some oracle rac installations but those I don't consider real clusters either as the database file system clustering is not general purpose like Luster or GFS.

blablabla

[Otis_INF]

Windows has proven time and again to be designed for stand-alone situation. All network and security add-ons have shown to be just that; add-ons..
huh? COM+ is designed to be a cross-machine/process object layer with security build in PER OBJECT, even per interface. Role based, AD controlled.

Stand alone? Add-ons? ever looked closely at windows 2000 or even NT 4? No, not the shell, the core OS.

Distributed computing simply isn't part of the base design. Morphing Windows into something it isn't will once again be a task for their marketing department, not engineering.
You have definitely totally no clue whatsoever, and with you the moderators who modded you 'insightful'. 'Bullshit' would have been more like it.

Re:blablabla

[Goeland86]

I'm afraid that this time, you're not looking at where MS comes from. He does have a point! Windows, yes, even NT, evolved from the same point. They wanted to start from scratch, so re-wrote alot of the code. But there are so many flaws common to both NT-based and DOS-based versions of windows that you can assume they didn't rewrite the entire OS, only the parts of it that were preventing a good network management. Now look also at the licensing of windows: each workstation has to run it's own OS, and for each workstation you have to pay a license. So, yes, indeed, even NT, 2000 or XP is a standalone OS. It's not the technology that's lacking to have multiple terminals from the same kernel, in fact, Microsoft does have them, they just don't advertise it. Just like the LTSP project, Microsoft has it's own server/client software. So why is it that every company seems to need each employee to use his or her own computer as a standalone? And if Windows was even partially designed for distributed computing I'd be interested to know why no one has attempted this before? Surely you can't tell me someone at MS wouldn't have thought about it until just now... I for one welcome the fact that MS is trying to enter a market where linux dominates, because in my opinion it'll show Windows' weaknesses, and more importantly, that MS is a very good marketing company, except that their software isn't up to the challenge.

since when are programs ran when they're not used?

[Otis_INF]

in Windows every processor would also have to run the entire GUI. Even if it is never used.
No. First of all you can set cmd.exe as the shell instead of explorer.exe, second of all, if you don't hook up a monitor or log in, the shell is swapped out pretty fast, and doesn't get any cpu cycles.

clustered filesystem

[Otis_INF]

ah, you never saw a rack of windows servers access a separate file cabinet?

But can it run from a cd?

[codepunk]

http://bofh.be/clusterknoppix/

Old BirdDog fetches dead bird

[xtermin8]

I suspect MS is looking for a way to squeeze some extra money out of this, perhaps implement some subscription services which they've been trying to do for a while. There would be a lot of smirks though, if IBM and Apple enter an unholy alliance and corner the low-end clustering market.

Re:since when are programs ran when they're not us

[peragrin]

Okay that does answer one question I have had.

Now for the second.

Are their DOS only based utilities to edit registry settings? since everything in windows is configured from the registry how do you edit those settings?

How do you change the hardware settings?

Hello! Editiors, are you awake?

[DAldredge]

A lot is TWO WORDS. Why is it so hard for you to do the jobs you are paid to do?

Re:since when are programs ran when they're not us

[Otis_INF]

Are their DOS only based utilities to edit registry settings? since everything in windows is configured from the registry how do you edit those settings?
Yes, they're in the resourcekit for windows 2000/2003/XP (and a lot of other command line tools)

But you don't have to do this, you can for example remotely login using terminal services for admin usage, even if the server doesn't have monitor,mouse and keyboard attached. But if you want to config windows using a commandline, you can.

rescanning hardware changes can be triggered by a reboot.

Would it be an addition to home users computers?

[ElQueso]

Would it be an addition to home users computers as well as the server versions of Windows?

Please, home users don't even get a decent web server, just a crippled version of IIS, let alone "clustering"

Most likely they are going to release a special version "Microsoft Windows 2023 distributed datacenter edition", and the version number is going to be something like "Windows NT 4 [Version 7.00.3879]"

Research grid systems are mostly *nix

[Danathar]

I think MS has finally admitted that until windoze can match *nix (LINUX, UNIX, OS X) in the distributed computing sphere, research is not going to touch their stuff.

Fact of the matter is they have a pretty hard uphill battle ahead of them. The research computing community is as pro-linux and UNIX as any zealot here on slashdot.

Nearly the entire U.S. goverment uses UNIX (mostly LINUX actually) within the supercomputing realm. DOE and NSF's supercomputing centers all run LINUX.

We'll know how serious they are by their presence at the next supercomputing conference.

-Doug

Windows copies zombie spammer code

[relaxrelax]

But aren't the malware zombie windows hackers having a patent on this code? Wouldn't they sue?

Microsoft Cluster Server (MSCS)

[not_hylas( )]

Ahem, (tap, tap, tap):

Clustering Solutions for Windows NT:

http://www.windowsitpro.com/Windows/Article/Arti cl eID/228/228.html

http://www.amazon.com/exec/obidos/tg/detail/-/01 30 960195/102-3088378-9911361?v=glance

http://research.microsoft.com/users/gray/Wolfpac k_ Compcon.doc

I can't be the only one that had this book.

Re:Microsoft Cluster Server (MSCS)

[iggymanz]

it seems this article is talking about a more loosely distributed computing model than clustering, but then I'd say Windows has been capable of many types of distributed and parallel architectures for years too - with middleware, with various rpc, with many tools that are used in the Linux/Unix world to build both distributed and parallel systems also ported to windows (such as many flavors of MPI. Just another way to do the same stuff that's already been possible for at least 5 years...

No comparison

[burnin1965]

I still feel that Linux would be a good bit cheaper, but we're talking big bucks both ways.

Perhaps the lack of detail in your arguement has given me the wrong impression, but it sure looks to me like you or bordering on soft fud. If I misrepresent your arguement then I appoligize.

You are likely refering to customization as rewriting kernel code or writing new code for inclusion in the kernel which would be expensive either way (I suspect much more expensive for Windows because you would be paying for more than just software development, MS would need their pound of flesh). However, the parent you replied to was talking about customizing the feature and capability set compiled into the kernel for a specific application.

The linux kernel source and development tools are designed with this flexibility in mind and for the majority of applications I suspect an end user would not require custom kernel coding but could run the kernel configuration and recompile with minimal to zero cost. If you go through a 2.6 kernel configuration you will find that compiling for a specific processor is only one of several hundred customizable features.

A linux kernel can be easily customized as a fully featured kernel for operation on a high powered workstation with a wide array of devices and services, to a highly specific and efficient server kernel for a specific service application such as web services, packet routing and filtering, file and application serving, etc., all the way down to a bare minimum kernel which will be served rapidly and efficiently over a network for a diskless distributed processing cluster node. Many of these easily accomplished tweaks in the linux kernel are trivial to accomplish but are not trivial in what they actually accomplish.

Although for some outrageous sum of money and signing of contracts and licenses to protect corporate "IP" I'm sure somebody could pull off the same customization of the NT kernel, however, I would still argue that the NT kernel was not designed for this type of customization and it is nowhere near as inexpensive or as simple as it is with the linux kernel.

Making custom NT kernels for any application is not trivial while custom linux kernels for most applications are trivial. There is no comparison between the two.


Aside from the disagreement I suspect nothing significant will come of this latest MS effort and it is more of an attempt to have their corporate fingers in every game. There has been lots of press lately about large linux clusters and I think MS just wants to be able to say "oh yeah, look me too".

burnin

Re:On the other hand...

[cammoblammo]

Morphing Linux into something it isn't will once again be a task for a commercial corporation, not OSS programmers.

Since when have commercial corporations and OSS programmers been mutually exclusive?

Re:Not So Funny: China and its Threat

[Anonymous Coward]

I'm starting to fall in love with you, China troll.

Re:Been done already

[myukew]

Why is the parent a Troll whereas like ten similar posts are modded "funny" or "insightful"?

Do I really need to understand the modding system?
Probably I'm modded Troll myself for this...

BRING BACK THE TROLLS!!!

[mangu]

Flamebait? Please mods, this is the reason slashdot is losing readership.

No, I don't think that's the reason. Metamod takes care of clueless mods. I don't know about others, but the main reason why I read much less Slashdot than I used to is that trolls have been effectively defanged.

Slashdot was a really funny site to read when one could find humorous, although often off-topic, gems of internet wisdom inside. Most of the posts that get modded "funny" today would be better classified as "trite" or "corny". Today, the people who made Slashdot what it was have been banned. No more Natalie Portman, no more goatse ascii, no more anything that's even remotely diferent from the mainstream media.

Well, for mainstream media we have Geraldo and Larry King, thanks, we don't need Slashdot for that. For discussions on current technology news there are gadzillions of sites in the net. What made Slashdot truly unique and fun to read were the trolls, and they have been effectively eliminated.

The troll-elimination effort has gone way too far, it's harming the technical discussions. Last week I tried to post a small snippet of perl code in my comment. It was rejected by the lameness filter, because it looked "too much like ascii art"!!!... Well, for mature and well-balanced technical discussions, posting the occasional five-line perl code is invaluable. If you don't want to see the occasional ascii-goatse, you *do* have some ways to protect your sensitive retinas, did you know that, editors?

Well, yes, I agree that the "flamebait" moderation hurts Slashdot. So, here is my own constructive proposal: when giving mod points, let only *ONE* point be negative. You get five points, you must mod four posts up, and only one post down. So you must choose really carefully which one is a troll or flamebait or overrated or redundant. It's not only the excess of flamebait moderations that's bringing Slashdot down, it's a general excess of negative moderation.

Re: BRING BACK THE TROLLS!!!

[SunFan]

ascii-goatse

The goatse art trolls do provide a rather expressive use of their medium, and it did take quite a bit of effort to refine their character usage to properly capture the essense of their still life subject.

hmm not 4 me..

[josepha48]

I don't think, given their current virus propagation, I'd want a grid of windows boxes. I'd guess one gets a virus and it would spread to the grid. They need to resolve the security issues first and then they can do the grid computing. I'd say we wont see this working effectively for at least 5 years. Yes they may have it "working" now, but I'd rather run a grid of *BSD's / *NIX than Windows. Just my 2cents.

Re:you are obviously a paid microsoft shill

[heffrey]

Ooh, that's nasty. And you are.... That's right, Anonymous Coward! Not paid by Microsoft actually. Just capable of independent thought.

Re:Been done already

[AtariAmarok]

Actually, it should have been modded down "redundant". After I posted, I saw that there were already a couple like what I wrote. I violated the posting rule "Read other people's messages before posting your own to avoid simply duplicating what has already been said." Tsk tsk tsk.

But you do have a point. Since when is Microsoft-bashing considered to be "Trolling" on Slashdot?

Re:Written in C#?

[Duhavid]

Because there is a marketing purpose to be served here.

Alittle petpeeve

[uncadonna]

Wheredo peopleget the ideathat alot is aword?

Could they change that idea, please?

Re:Alittle petpeeve

[rogerborn]

uncadonna:

The same place *their* getting *there* words 'their' 'they're' and 'there' mixed up. =)

Seems it's not only the intelligence level of /.'ers that is falling lately, but their language is devolving too.

Or perhaps it is the great temporary influx of Windows fans appearing in this particular blog.

Regards, and Happy New Year!
Roger Born
Writer, Teacher, General Troublemaker
writing.borngraphics.com
"Sorry, no refunds."

Why did I read this headline as

[Flyboy Connor]

Why did I read this headline as:

"Microsoft Finally up for Disturbed Computing?"

Parallel?

[LanceUppercut]

Look like the guy doesn't understand the difference between "distributed" and "parallel" computing. Multithreaded parallel computing was a native feature of all Widnows NT systems from the very beginning.

GRID Computing With Microsoft...

[hackus]

WoW!!!!

I hope the licensing server and your bank account can scale along with the computing requirements!!!

-Hack

Re:since when are programs ran when they're not us

[m_pll]

Reg.exe is shipped as part of the OS starting with XP.

Feh...nothing new

[Dr. Sigmund Freud]

... with all the zombie PCs out there, Windows has proved its ability to do distributed computing for quite a while.

some interesting docs

[plopez]

a couple of interesting docs pop up when you google this:
+"grid computing" site:microsoft.com

Globus

[nr]

Would really like to see Microsoft implement the new Globus 4 in .NET and C# so they can play along with the big boys. The new rewritten Globus 4 standard is build up on W3C standards like SOAP/WSDL and other WS standards so Windows machines can now become equal Grid nodes.

www.globus.org

who would buy this

[suezz]

who would actually buy this - anybody who does this computing is strictly a unix shop - but I guess Marketing needs to put together their latest powerpoint presentation and it can't be done on one computer anymore. great - problems have just multilplied.

attention idiot linux bigots:

[nerdb0t]

microsoft systems have done parallel computing just fine for many years.

in 1998 a 16 node NT system at UC Berkeley broke the worlds record for parallel sorting (the previous record was held by a solaris cluster.)

parallel microsoft systems work very well - and note there is nothing *magic* about the linux operating system and parallel systems. it's just an operating system - almost all parallel computing is done in user space.

check your facts you freaking bigotted morons. i dont know why i bother reading the comments on slashdot - so many idiots, but in this case even the article submission was in error.

in 2 years

[geekoid]

they'll tlak like they invented it.
Just like DOS, GUI interface, the internet etc . . .

maybe he just

[geekoid]

misspelled 'plot'?

heh sore but there are alot of people whining about this thing alot.

Re:standard distribution

[Doc Ruby]

Why are you so freaked out by the tiny barb associating Microsoft and their buggy, security threat OS with money? They've got *your* money, too - and we're all in this distributed Windows mess together. What's funny about that?

Re:standard distribution

[Doc Ruby]

You're the one calling it "funny", bringing up the Not Funny _User Friendly_, and using "uber-". If you don't like it, change the channel.

Re:Not So Funny: China and its Threat

[jack_csk]

Insightful your ass!

Even without Windows, China is already doing grid computing. Red Grid is an example. Given that the China goverment prefer Linux than Microsoft Windows,they are already using clusters to do whatever they want.

If somebody mod me down, he/she must be as clueless as Mr. / Ms. Coward.

isn't it already done?

[Hackysack]

when you need to launch an internet wide DOS attack against a particular website, what do you use? slaved/zombied winXX boxes, you've hijacked.

When you need to send spam on a global scale, go for the zombified, comprimised winXX boxes again!

I thought XP/2k/ME was distributed computing, I mean anyone can run anything on anyone elses box.

MPI works for windows

[S3D]

Considering that the Windows platform has never had the ability to parallel compute in the past, it leaves great potential to the company's operating system development. That is not entirely correct. An MPI is working for windows and allow to build really working windows cluster. Me and another person made such a cluster several years ago, it was more difficalt then Linux cluster, but it was working. (Why we did it - we were building cluster for 3d rendering and had to use DirectX API)

Parallel Computing it is...

[ReeprFlame]

My bad. It did even say it right in the article. Guess I better fix that coffee machine so I am awake next time I write the article...

May we stop bashing M$ and come up for some air?

[the angry liberal]

ReeprFlame writes "eWeek has reported overhearing Microsoft's plans to finally get into the distributed computing market. Considering that the Windows platform has never had the ability to parallel compute in the past, it leaves great potential to the company's operating system development. From current *nix systems we have today, such a grid proves very useful, especially in the serving arena. However, we are unsure of Microsoft's target for the software. Would it be an addition to home users computers as well as the server versions of Windows? As of now it is unclear, but Microsoft probably will bring this situation to life in the near future since it does hold alot of power for them over other platforms."

I arrived a bit late on the scene to advise you guys, but here goes: If you are going to have a derogatory thread about M$ and how lame they are, um, get your terminology right first!

Parallel =! Distributed
Distributed =! Parallel

These are two different things and demonstrates how this board has a tendancy to go off a bit half-cocked over what M$ does and their capabilities.

What is even sadder, those speaking seriously of problems with Windows 95 (BSoD) having anything remotely to do with enterprise level systems today. Being an old-timer BSD and Linux user, I can safely say I've seen a share of pings of death and other BS with every OS. Sometimes, it becomes time to just let all that childish BS go, ya know? Implement what businesses need and works.

Re:Been done already

[myukew]

Guess Microsoft infiltrated the modding system...