Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

openMosix Is Shutting Down

kdawson posted more than 7 years ago | from the who-needs-clusters-anyway dept.

Software 252

jd writes "Despite having one of the largest user-bases of any clustering system for Linux, openMosix is to be shut down. Top developers have left and they lack the means or motivation to continue. Their official claim of multicore CPUs making clustering redundant is somewhere between highly improbable and totally absurd, as has been pointed out elsewhere. Why is this shutdown so important? Well, from a technical standpoint, the open-source bproc (the Beowulf process migration module) is ancient, MOSIX is very hard to obtain unless you're a student, and kerrighd is (as yet) immature. From a user standpoint, openMosix is the mainstay of the Open Source clustering world and has by far the best management tools of any. The ability of this project to continue will likely have a major impact on the future of Open Source in the high-end markets — if the best of the best couldn't survive, people will be more careful about anything less."

cancel ×

252 comments

Sorry! There are no comments related to the filter you selected.

orly? (2, Interesting)

xednieht (1117791) | more than 7 years ago | (#19896127)

If they were the best of the best of the best they would not be shutting down.... The best of the best find a way, and when they're done they go home and show the prom queen the difference between ROM and RAM!!

Re:orly? (1)

thatskinnyguy (1129515) | more than 7 years ago | (#19896183)

Well, Mosix did what it did very well. Now there are alternatives to do what Mosix did without all the power, space and hardware considerations.

Re:orly? (-1, Offtopic)

Anonymous Coward | more than 7 years ago | (#19896991)

Do you know what else is shutting down? My balls. I'm having them removed next week so my penis can be made into a vagina.

Re:orly? (-1, Troll)

Anonymous Coward | more than 7 years ago | (#19896535)

I had a beowulf cluster of prom queens once. Unfortunately, the great state of California considered that a crime (16 will get you 20) and now my sex life consists of being shower raped by niggers and mexicans. My lawyer has filed a constitutional lawsuit (on the grounds that being infected with HIV via prison rape is violates the fourth amendment) but even if it succeeds, my life is over.

Re:orly? (0)

Anonymous Coward | more than 7 years ago | (#19897207)

God bless you sir, I trust that your lawyer will free you to once again frolic with prom queens!

Re:orly? (1)

Ruie (30480) | more than 7 years ago | (#19896557)

From a user standpoint, openMosix is the mainstay of the Open Source clustering world ...

Sorry, but I have never worked on an openMosix cluster, but did use several condor-based ones. Also, the NSF-backed grid project is targeted at large and relatively slow networks of computers rather than an emulation of a NUMA-like architecture.

So, while Mosix ideas are certainly cool, etc, it is not a mainstay of anything.

Also, what is it with "MOSIX is very hard to obtain unless you're a student" - is this the same kind of "open" as in OpenXML ?

Re:orly? (1, Informative)

kwalker (1383) | more than 7 years ago | (#19896733)

MOSIX [mosix.org] and OpenMOSIX are separate projects. OpenMOSIX was started because MOSIX is so closed. Try to find an easy download link for MOSIX, go ahead. I'll wait...

I'm saddened by this development too. I've got a small network I've built over years of tinkering with Linux and I would have liked to explore what MOSIX and OpenMOSIX promised. I was hopeful that OpenMOSIX would release a stable branch for Linux 2.6, as that's what I prefer running on my machines. I may have even been able to contribute some after a while, but I'm no kernel hacker (Which is what's required for a project like this), so I can't even bootstrap in now.

Re:orly? (1)

Ruie (30480) | more than 7 years ago | (#19896807)

MOSIX and OpenMOSIX are separate projects. OpenMOSIX was started because MOSIX is so closed. Try to find an easy download link for MOSIX, go ahead. I'll wait...

Ahh, this clears it up - thank you ! (and sorry to the OpenMosix folks).

Re:orly? (1, Informative)

thejynxed (831517) | more than 7 years ago | (#19897057)

You mean links like these?

http://linux.softpedia.com/get/System/Operating-Sy stems/Kernels/MOSIX-7287.shtml [softpedia.com]
http://linux.softpedia.com/get/System/Operating-Sy stems/Kernels/MOSIX-Grid-and-Cluster-Management-23 125.shtml [softpedia.com]
http://www.mosix.org/txt_cluster.html [mosix.org]
http://www.tucows.com/software_detail.html?id=8473 [tucows.com]
http://www.icewalkers.com/Linux/Software/530140/MO SIX.html [icewalkers.com]

BTW, that's just a few. I hope they helped out. BTW, my search term in Google were "MOSIX download" without the quotation marks.

Re:orly? (3, Insightful)

Thomas the Doubter (1016806) | more than 7 years ago | (#19896597)

I have been following openMosix for many years and, to be honest, the development team was never much good. OpenMosix worked best shortly after it was forked from Mosix, and it was down-hill ever since - about 10 years now! The latest stuff for the 2.6 kernel never worked at all. That said, this cluster-level kernel-hacking is probably just too hard to do right, unless you were to start from the ground-up. Plan9 anyone?

YA, RLY. (-1, Troll)

Anonymous Coward | more than 7 years ago | (#19897279)

And this, dear FOSSies, is why the enterprise will never choose Lunix.

If you cannot provide an enterprise quality product, enterprise quality support, and enterprise quality reliability... nobody is going to view Lunix as anything more than a kid's tinkerer tech toy. Which is essentially what Lunix is.

Anyone who talked their corporation into choosing a Lunix cluster better start polishing up their resume. Maybe "the community" can help you after you get shit-canned. BWAHAHAHAA!! Losers.

But hey, take heart: it could be worse. You could be the IT staff working for Munich, Germany, which has been beating their head against the wall for five years trying to get everything moved onto teh Lunix. Imagine being responsible for bringing an entire municipal government to a virtual standstill because of a horrible decision like that! And all because of some slavish devotion to creating a Lunix monoculture.

If it's really necessary... (4, Insightful)

42Penguins (861511) | more than 7 years ago | (#19896139)

someone else will pick it up.
Isn't that kind of the point of open source?

Re:If it's really necessary... (5, Insightful)

ozphx (1061292) | more than 7 years ago | (#19896207)

That is the theory of open-source. In practice the set of core contributors to a project are its foundation. As these people are leaving it will be extremely difficult to find others with the knowledge and motivation to continue its maintainance.

As with any project requiring something a lot more than a hobbyist the level of expertise required to work on the codebase is rare, and not cheap.

The only real hope is that a company or university using it is happy to pick up the tab and pay someone.

Unfortunately the "everyone can see the source code" line doesnt give any comfort when you are talking specialised things like clustering. I probably know a total of one person with the skill to work on such a system, and last I spoke to him he was contracting at 130 an hour - for comparitively easy (and less stressful) .net/c# work.

Re:If it's really necessary... (-1, Troll)

Frosty Piss (770223) | more than 7 years ago | (#19896291)

As with any project requiring something a lot more than a hobbyist the level of expertise required to work on the codebase is rare, and not cheap.

So... Open Source really isn't the opium that RMS would like us to believe?

Re:If it's really necessary... (1)

Mr. Slippery (47854) | more than 7 years ago | (#19896715)

Open Source really isn't the opium that RMS would like us to believe?

RMS speaks about Free Software, not Open Source; and he has no problem with people being paid to work on, and selling, Free Software.

Re:If it's really necessary... (1)

hardburn (141468) | more than 7 years ago | (#19896731)

Using "open source" and "RMS" in the same sentence shows you don't really know much about RMS.

Re:If it's really necessary... (0)

LingNoi (1066278) | more than 7 years ago | (#19896379)

That is the theory of open-source.
Let me stop you right there. I have personally myself taken up open source applications where they were abandoned by the core developers.

well you aren't in that line of work (2, Insightful)

r00t (33219) | more than 7 years ago | (#19896601)

I could maintain it. I have 7 to 10 coworkers who could maintain it. At my previous place there was one guy who could have maintained it. At the place before that, there were over a dozen people who could have maintained it.

I would in fact maintain it if I cared. I don't care.

BTW, I have doubts about the .net/c# guy you know. Most people who could maintain Mosix would not tolerate such work. They'd look down on it like a typical C++ developer looks down on HTML or Visual Basic development.

Re:well you aren't in that line of work (4, Insightful)

bladesjester (774793) | more than 7 years ago | (#19896777)

They'd look down on it like a typical C++ developer looks down on HTML or Visual Basic development.

Yes, I'm sure they'd look down on a very well paying job that was far far less stressful.

Re:well you aren't in that line of work (3, Insightful)

IWannaBeAnAC (653701) | more than 7 years ago | (#19897165)

Since when was writing C++ code stressful? Surely if anything, writing VB code is stressful?!?!

Re:well you aren't in that line of work (3, Insightful)

bladesjester (774793) | more than 7 years ago | (#19897285)

Depends on the VB project. A lot of places that actually want it done in VB just don't have someone around who codes and want something worked on. Unless it's a rat's nest, it's usually not so bad, and at $130/hour, I'm sure you could cope.

Besides, the person I initially replied to was of the opinion that nobody who could do the sort of work that openMosix requires would deign to "dumb himself down" (figuratively speaking) to writing c#/.net code even if it was netting the guy $130/hr. Personally, I call BS. After a while, you learn that you work to live instead of living to work.

Re:well you aren't in that line of work (5, Insightful)

r00t (33219) | more than 7 years ago | (#19897335)

The really good hackers:

a. don't want their minds and skills to rot
b. get bored by the easy stuff
c. are not stressed by difficult hacking (stress comes from office politics)
d. like to be admired for their ability to do the difficult stuff
e. like to be in the company of peers who can do the difficult stuff

You might get a great hacker doing lame stuff, but you'd have to pay him much MORE than you'd have to pay him to do the difficult stuff. The extra pay would compensate for the extra boredom. Since you can get a warm body for much less money, you're unlikly to hire the great hacker.

Since C#/.net is very lame compared to the challenges of something like OpenMosix, we can pretty reliably conclude that the supposed hacker is not really qualified to hack on OpenMosix. (alternate theory: his dad is the CEO and so the pay is quite absurd for the job being done)

Re:well you aren't in that line of work (5, Insightful)

bladesjester (774793) | more than 7 years ago | (#19897521)

Let me let you in on a little secret. Even the best people eventually realize that there's more to life than working no matter how "cool" you think what they're working on is. They look at their lives and realize that living to work is a bad idea because life is for actually living.

For a lot of people, that happens about the time they have their first kid. For others, it happens sooner. Yet others experience it later, to the detriment of their families if they have them.

I also have to tell you that it's not uncommon for a good independant contractor to be paid more than $130/hour because most consulting companies bill out their contractors at that much or more. Honestly speaking, my top hourly rate thus far has been more than $130/hr.

You may learn that your ideal of the "great hacker" is rather off the mark some day. The truth is that the really good people often don't care about how great others think they are. They get things done, and move on with what they have to do.

Re:If it's really necessary... (4, Insightful)

bladesjester (774793) | more than 7 years ago | (#19896209)

someone else will pick it up.
Isn't that kind of the point of open source?


It's a nice theory, but it doesn't really work out that way. If the lead devs leave a large project, the task of other people getting up to speed can be huge to impossible. It takes a long time to learn a system, especially if you're just doing it as a hobby.

Brain drain is a problem in any project, open or closed.

Re:If it's really necessary... (2, Insightful)

Frosty Piss (770223) | more than 7 years ago | (#19896259)

It's a nice theory, but it doesn't really work out that way.

Well, if no one picks it up, than clearly, it's not as popular as we are led to believe. Honestly, it it's that impotent, development will continue, otherwise, maybe it's not that important.

Re:If it's really necessary... (1)

kestasjk (933987) | more than 7 years ago | (#19897153)

This is where comments and documentation become vital. If a project is popular, logically built and well commented maintenance is possible without needing intimate knowledge, and once you have maintainers it's a small step to developers.

Re:If it's really necessary... (1)

bladesjester (774793) | more than 7 years ago | (#19897255)

A large program is difficult to get up to speed on, comments or no comments. Good comments certainly make it a lot easier, but it's still no small task. It takes a long time to get familiar with large systems to the point that you don't have to spend 90% of your time trying to find the place that you're looking for.

Re:If it's really necessary... (0, Offtopic)

70Bang (805280) | more than 7 years ago | (#19896221)

"someone else will pick it up"
and
"MOSIX is very hard to obtain unless you're a student

__________________________________________________ __

Both of these things can best be solved by an obvious solution.

Skill retention is not easy (2, Interesting)

EmbeddedJanitor (597831) | more than 7 years ago | (#19896275)

Having access to the source is one thing. Keeping the best people engaged in those projects is quite another.

There are many very valuable projects that get very little funding - insufficient to pay the programmers who give that value. If the contributors cannot live by their work then they have to go find payment elsewhere.

As open source matures, people will come to understand that taking without giving back is not a sustainable model.

Looks like abandonware (0)

Anonymous Coward | more than 7 years ago | (#19896417)

Even the license isn't clear. It seems to be GPL, but CVS commits have lots of comments like OpenMosix is no longer free software. What does that mean? I recall their was an ownership rift 5 years ago. I wonder if they ever got OpenMosix running without the non-GPL code.

(my slashdot human verification image was "abandons" - how fitting)

Re:If it's really necessary... (0)

Anonymous Coward | more than 7 years ago | (#19897235)

Sure, all that someone will have to do is look at the excellent documentation that describes the architecture and design philosophy. Oh? That doesn't exist because it's more fun to code new stuff than grind out good documentation that actually explains things (as opposed to necessary, but not sufficient, stuff that defines what modules do and what the API calls are?)

Sad (1)

thatskinnyguy (1129515) | more than 7 years ago | (#19896151)

I read about that yesterday. I had a lot of good times with it. I thought it was fun to create ad-hoc Beowulf clusters afterhours in the computer labs at the school I used to work at with clusterknoppix. The Mosix website had a valid point about multi-core CPUs and VMs. OpenMosix... RIP

OpenSSI (4, Informative)

sonamchauhan (587356) | more than 7 years ago | (#19896155)

There's a similar project named 'Open Single System Image'

http://sourceforge.net/projects/ssic-linux [sourceforge.net]

Re:OpenSSI (2, Informative)

TheRealMindChild (743925) | more than 7 years ago | (#19896485)

ANd I cant claim that I know what I am talking about, but it seems that OpenMosix was akin to Beowulf clustering... wherein the application needs to be aware (or a library abstraction) that it is working in a cluster.

OpenSSI, however, works transparently to the userland, at least. Got a program that spawns bunches of processes? OpenSSI.

But don't get too excited. This is NOT exactly like SMP. Unlike multiple physical processors that can actually have threads of x application running on different procs, OpenSSI only migrates processes. So your super multithreaded game wont magically work better.

Re:OpenSSI (3, Interesting)

jd (1658) | more than 7 years ago | (#19896985)

OpenMosix was also fork-and-forget at the PID level. There was an effort to make it fork-and-forget at the level of individual threads, but nobody could figure out how to solve the latency hell that is synchronized shared memory. I believe that it may be partially solvable by using reliable multicasts - only one transmit per update, not one transmit per node - and by using kernel bypass tricks to avoid the 20ms context switch for large updates.

OpenSSI was part of one-stop solutions, if I remember correctly, the doomed Compaq foray into clustering before HP took them over. Doomed? Well, HP has not exactly been Linux-friendly. Their efforts to be more so by hiring Bruce Perens never panned out and you certainly don't see them porting any of their HPUX security to Linux.

I for one (0)

Anonymous Coward | more than 7 years ago | (#19896159)

I for one welcome our new non-OpenMosix overlords.

Re:I for one (2, Funny)

Anonymous Coward | more than 7 years ago | (#19896233)

Just imagine a beowulf clus... Oh never mind.

Re:I for one (0)

Anonymous Coward | more than 7 years ago | (#19896285)

All your openMosix are belong to us.

No ulterior motive or competing interest then... (4, Interesting)

Anonymous Coward | more than 7 years ago | (#19896211)

FTA: "Moshe Bar, openMosix founder and project leader, has announced plans to end the openMosix Project effective March 1, 2008."

Wikipedia: Moshe is founder of the company behind the Xen software, XenSource, Inc. Moshe is also founder of the company Qumranet which is behind the development of the KVM virtualization technology in the Linux kernel.

Looks like Moshe is to busy for that old fashioned mosix stuff...

Huh? (3, Funny)

wilymage (934907) | more than 7 years ago | (#19896225)

"[T]he open-source world progresses with giant steps. It is a world where the sun never sets and where national borders, race and religion have no meaning. What counts is the code. And that comes abundantly, and in high quality." (attributed to Moshe Bar on his site)

Apparently the code doesn't count, only spurious logic about changing hardware factors. Oh, and apparently the sun does, in fact, set.

But how cool a name is Moshe Bar?

Uh, I think the summary misses the point of OSS... (4, Insightful)

WIAKywbfatw (307557) | more than 7 years ago | (#19896245)

From the summary it seems that the people who've contributed most to the openMosix code have moved on to other things.

Well, that happens. People's lives don't stand still, they change: they take on other commitments at work, have relationships, travel the world, etc.

But that doesn't mean that openMosix is dead.

On the contrary. This is open source software.

The code isn't lost. Others can pick up the slack and join the effort as they see fit. openMosix can still move forward, perhaps not at the same pace as before, but forward nevertheless.

It seems to me that the summary misses the point of OSS. If this was a closed source project and the lead developers had walked away then, yes, openMosix would almost certainly be dead and buried.

But, unless I'm missing something huge this isn't the end of the line for openMosix, precisely because it is open source.

It hardly seems appropriate to look at this as a failing of OSS development. On the contrary, it's arguably an example of one of its strengths.

This a baton change not a retirement. At best, the new holder(s) of the baton will soon hit the same stride as the previous holder(s). At worst, the baton has fallen to the ground and it simply needs to be picked up.

Re:Uh, I think the summary misses the point of OSS (1)

Maniac-X (825402) | more than 7 years ago | (#19896283)

Unfortunately not everyone is as optimistic about it as you, but I hope that they're wrong ;)

Re:Uh, I think the summary misses the point of OSS (3, Interesting)

DogDude (805747) | more than 7 years ago | (#19896289)

Yeah, but how likely or realistic is it that the few people in the world who understand, in this case, clustering, to such an extent, will choose to work on this project? The vast majority of software developers want to get paid for their work.

In theory, you're right. It'll continue. But will reality live up to theory? Only time will tell.

Re:Uh, I think the summary misses the point of OSS (4, Informative)

WIAKywbfatw (307557) | more than 7 years ago | (#19896385)

I agree, there's a degree of optimism in my argument but the summary is plain flawed.

Its message and tone is that openMosix = dead, openMosix = OSS, therefore openMosix dying = OSS solutions are bad.

What it completely fails to address is that the situation would be no better, and in fact would be a lot worse, if this was a CSS tool. Indeed, the ray of light for openMosix users comes from the fact that it is OSS.

Bashing OSS solutions because one is dead/dying/in limbo/whichever way you want to look at it is patently ridiculous because it's not the openness of the code that's at fault here, or even the open source development model.

To put it bluntly, CSS projects that lose their core development teams don't exactly fair any better do they?

Re:Uh, I think the summary misses the point of OSS (1)

Telvin_3d (855514) | more than 7 years ago | (#19896611)

Yes and no. CSS at this scale always means commercial. One thing about commercial software is that popular, well used competitive software will not disappear because people got bored and moved onto other things. When the development of a piece of software provides a revenue stream, it is not going anywhere. People will be hired, an interested competitor will buy it out in order to give themselves a leg up, whatever. There are all sorts of survival mechanisms available to commercial software that doesn't exist for OSS. Sure, there is lots of abandoned CSS out there, and it is sad that the code is not available to the few faithful that would want it. However, CSS that is as large in its speciality as this seems to be doesn't. It might fade if something better comes along, but it won't just disappear when things are going good.

Re:Uh, I think the summary misses the point of OSS (4, Insightful)

rtb61 (674572) | more than 7 years ago | (#19896849)

Numerous closed source projects are killed all of the time and for all sorts of reasons. For example bought out by a competitor and then just killed regardless of user base desirability and all of it's paid contributors fired on the sport all to eliminate competition, or simply die on the vine, not because of bad code or poor programmers or even a lack of users, just bloated management bleeding a company dry until it fails or killed by a company making use of monopolistic tactics.

Some utility bits of open source of course do not need a lot of maintaining and reach full maturity pretty early and only require the odd tweak for hardware compatibility, for those projects maintaining a team is difficult, logically speaking those projects get pick up and carried by another open source project that can run them as a side line.

Re:Uh, I think the summary misses the point of OSS (1)

Fred Ferrigno (122319) | more than 7 years ago | (#19897323)

One thing about commercial software is that popular, well used competitive software will not disappear because people got bored and moved onto other things.
Popular, well-used open source software won't disappear either, because somebody else will take over. If no one steps up, then it wasn't really that popular to begin with. Or if some company really needs the software badly, they can hire someone to work on it.

Re:Uh, I think the summary misses the point of OSS (1)

Max Littlemore (1001285) | more than 7 years ago | (#19896705)

To put it bluntly, CSS projects that lose their core development teams don't exactly fair any better do they?

On the contrary. A CSS product could be rebadged as openMosix 2008 Horizons(R), given some fancy UI tools and sold for more than double to a bunch of mums and dads who believe they need a cluster of 4 new core duo PCs to run a web browsing platform which fully enables their personal web experience for tomorrow today.

It's soooo much easier to make money from closed source software.

Re:Uh, I think the summary misses the point of OSS (1)

iamacat (583406) | more than 7 years ago | (#19897447)

You are actually making the opposite point from what you intended. Most families have multiple desktops and notebooks and could realistically benefit from pooling hardware for realtime home video editing, gaming, distributed backups or math/physics educational software. Yet, it takes geeks who only use OSS to actually set up clustering.

As for money part, we are talking about project abandoned by original developers. With proper legal support though, OSS would be easier to profit from than CSS as other people would create derivative works and pay you royalties.

Re:Uh, I think the summary misses the point of OSS (1)

Mr. Slippery (47854) | more than 7 years ago | (#19896637)

The vast majority of software developers want to get paid for their work.

And if this work is important, the community of users will organize and hire software developers to work on it. Thanks to the fact that it is Free Sotfware, the user comminity can do this.

Re:Uh, I think the summary misses the point of OSS (1)

larry bagina (561269) | more than 7 years ago | (#19896547)

look at sourceforge much?

"the mainstay of Open Source clustering..."? (2, Insightful)

Anonymous Coward | more than 7 years ago | (#19896325)

From a user standpoint, openMosix is the mainstay of the Open Source clustering world and has by far the best management tools of any.

LOL what?! The poster must be on crack. OpenMosix/Mosix is nothing but an experimental/buggy piece of software used by hobby clusterers, it works with 2.4 kernels but never had good support on 2.6. Real cluster software consists of PBS/Maui or some other queueing/scheduler built in house.

Article summary is an overreaction (4, Insightful)

Mag7 (69118) | more than 7 years ago | (#19896343)

What a ghastly overreaction, but hey, this is slashdot.

Best of the best? I may get flamed for this, but I'd barely heard of OpenMosix.

When Apache, the Linux kernel, Eclipse and (name a popular GNU project) look like "shutting down", then maybe we can bleat about the failure of open source.

And as some have said, there's not real reason the baton can't be passed on to interested new parties.

Re:Article summary is an overreaction (1)

rolfwind (528248) | more than 7 years ago | (#19896433)

Exactly. If it's open source and free, they could just fork it if it's that important, no? What is the bru-ha-ha about?

Re:Article summary is an overreaction (0)

Anonymous Coward | more than 7 years ago | (#19897289)

When Apache, the Linux kernel, Eclipse and (name a popular GNU project) look like "shutting down", then maybe we can bleat about the failure of open source.

Eclipse is already a failure for open source. What a piece of shit software.

But hey, it sounded great until I actually had to use it.

Re:Article summary is an overreaction (1)

moderatorrater (1095745) | more than 7 years ago | (#19897483)

Out of curiosity, how many clusters do you set up?

Linus is right (-1, Troll)

Anonymous Coward | more than 7 years ago | (#19896411)

I am with Linus on this one.
GPL v3 does not make sense. RMS should be stopped before it is too late.

Re:Linus is right (-1, Troll)

Anonymous Coward | more than 7 years ago | (#19897111)

I agree. GPL v3 is the solution to a problem that doesn't exist.

Why this IS important (3, Informative)

DFDumont (19326) | more than 7 years ago | (#19896413)

The pendulum has swung back now. In the days when 10Mbps ethernet came onto the scene and our processors could barely keep up with their floopy drives (which is why a floppy used DMA), we collectively came up with the idea of using several computers to solve a problem by splitting the problem up among them. Since then thanks to Moore's law our processors now spend a lot of time waiting to fetch the next instruction from their on-chip L1 cache - as in when there's a miscalculation in the branch prediction step.
Our networks however have not kept up to this pace. Right now our very best effort for network speed is infiniband which tops out at 96Gbps theoretical limit. The AMD Opteron page lists a limit of 24GBps, that's 192Gbps, bandwidth using three coherent hypertranport processors. See the problem?
I see one of two things happening, either we'll find a magic bullet technology to significantly increase our network speeds; or some limit will finally end Moore's law. Otherwise there's simply no reason to tie together multiple processors. Despite Windows best efforts, our CPU's still spend most of their time waiting for something to do.

Dennis Dumont

Re:Why this IS important (2, Informative)

1729 (581437) | more than 7 years ago | (#19896809)

I see one of two things happening, either we'll find a magic bullet technology to significantly increase our network speeds; or some limit will finally end Moore's law. Otherwise there's simply no reason to tie together multiple processors.
You might not have a need for a cluster, but that doesn't mean that nobody else needs them. We have quite a few of clusters where I work, ranging in size from about 4,000 processors to over 100,000 processors, and these machines aren't sitting idle. Multicore desktops systems are great for a lot of things, but if you want to tackle a really big, really difficult problem, a desktop system isn't going to cut it.

Re:Why this IS important (2, Insightful)

David Greene (463) | more than 7 years ago | (#19896975)

The moderators are woefully uninformed.

We are not at the end of network technology. You're talking about essentially consumer-level stuff. There's a vast amount of network technology out there that goes beyond what Infiniband provides.

Software that needs very high bandwidth won't work on a cluster and probably won't work very well on a single-socket desktop either. Right tool for the right job and all that. There are plenty of codes out there that want tens or hundreds of thousands of cores. Some can even run on clusters. Others need something a bit meatier.

It is telling Moshe Bar is now doing Xen and KVM (3, Insightful)

Ayanami Rei (621112) | more than 7 years ago | (#19897187)

Basically what we (the community) has figured out is that SSI featuresets should not be implemeneted in the OS layer, but below it. Look at the SGI Altix technology. Or large Unisys machines. Or that hyper-transport happy monstrosity that Cray is building. They have special low-level firmwares running on the I/O processors that are doing in low latency, tuned hardware what *Mosix was trying to do from Ring 0 on the nodes.

Using various ISA interfaces (MPI in the low end, or Hypervisor abstractions like Xen, etc. etc.) you can run many guest OSs in the space as needs require, and localize the shared-memory-ness as required to get maximum threading benefit with the lowest total latency you can tolerate. All this with minimally modified guest OSs in which to run the code. This is a much better situation then heavily modified kernels pretending to be a single system image (and then having to worry about forking/threading/VFS issues and propogation of that stuff).

On the flip side, grid technology and speciality message-passing libraries fill out the feature set for more embarassingly parallel problems that need lots of CPU and RAM... you have the luxury of spending time and money coding your applications for that environment if you are CPU limited.

Mosix doesn't have much use anymore as a general purpose product. Either it's too heavy-weight (and drowning in syncro overhead) and we should be relying on firmware/hypervisors that are customized for the hardware, or it's not necessary because we can handle the load balancing at a higher level.

Imagine (0)

Anonymous Coward | more than 7 years ago | (#19896427)

an openMosix cluster of these

Open Source Conundrum (-1, Troll)

Anonymous Coward | more than 7 years ago | (#19896429)

This is a textbook example of why business users are wary about open source. If a project can just dry up and blow away so easily, why would a company want to rely on it for their critical business needs? Sure projects like Linux and OpenOffice have reached a certain maturity and critical mass of developers and users that they are not likely to disappear overnight if at all. But you can bet your bottom dollar that Microsoft,Oracle,Sun etc. will not go away and leave you with a legacy piece of software that is no longer being actively being developed without providing for an alternate product that is viable over the long haul.

Re:Open Source Conundrum (3, Insightful)

pavera (320634) | more than 7 years ago | (#19896595)

Wrong.

MS regularly end of lifes things. Just recently the EOL'd foxpro. Sure its a crap language and a crap environment, but I know 5 people personally who are frantically trying to teach themselves .NET and get experience with that environment, as now that MS has declared foxpro dead, they aren't ever expecting to get another foxpro job. The only alternative left to companies with "legacy" foxpro code is to completely re-write any application in that language in a different one (not a small task).

You can still get the openMosix code, if you had openMosix experts you could still fix things and move forward. If you have an existing system on openMosix you can look for a different solution and move to it or keep your system on the existing code. I really don't see how this is any different than MS calling for an EOL of Windows NT. When they do that you are forced to invest tons of hours and money buying new systems, developing a migration plan, deploying the new system, training users on it... It is no different in Open source or closed source, when vendors decide they aren't supporting you anymore, it costs you money.

Vendors regularly leave users out in the cold, both closed source and open. Only difference is, if a company wanted to pick up openMosix they certainly could. They could provide support, ongoing development, whatever. When MS EOLs something, your only choice is to take whatever MS gives you.

Re:Open Source Conundrum (0)

Anonymous Coward | more than 7 years ago | (#19896743)

If they hadn't gotten to .NET by now, those Foxpro devs wouldn't have been able to fix Foxpro anyway and they're certainly not going spend their own money to pay somebody to maintain it. So what's your point?

Re:Open Source Conundrum (1)

timmarhy (659436) | more than 7 years ago | (#19896865)

"MS regularly end of lifes things"

yep, like how they EOL'd win98 last year after 9 years of support, and foxpro after 10+ years.

yeah their real bastards MS, only giving a product a 9 year life span. Those linux projects they just never EOL... oh wait no they don't they kill versions off after 2 years or less!

Face it. people have known the writing was on the wall for foxpro for atleast 3 years. they've had shitloads of time to learn .net. it's thier problem, not MS's

Not an open source 'problem' (1)

ToasterMonkey (467067) | more than 7 years ago | (#19896941)

What part of "It is provided "as is" without express or implied warranty." do wary business users not understand then? It's all up front. Not all open source projects are created equal, for sure, and many have a great deal of commercial support. To say businesses should be wary of relying the unpaid work of complete strangers with no contractual support is a big DUH isn't it? That goes for ANYTHING. That's why the Red Hats and Novells of the world exist.

Read between the lines people! (2, Insightful)

Secret Rabbit (914973) | more than 7 years ago | (#19896503)

"""
The direction of computing is clear and key developers are moving into newer virtualization approaches and other projects.
"""

Translation: The developers have found new shiny objects to play with and are going to drop this to play with something new.

Remember that OSS is mostly about developers scratching an itch. Once that itch is scratched, if a new shiny object is put in-front of a developer, chances are they'll drop what they're doing to pursue the new thing. As seems to be the case here.

i.e. New is fun, maintenance is boring, boring sucks, do something new.

Re:Read between the lines people! (1)

nuzak (959558) | more than 7 years ago | (#19896583)

Well, since no one in the small market for clusters was willing to throw significant cash at the problem, what other driver is there but developer interest?

Keep in mind that clustering in general isn't dead, just that much of it has moved to the language level. Parallelizing individual loops isn't something openMosix or anything else like it can do out of the box.

Re:Read between the lines people! (2, Insightful)

1729 (581437) | more than 7 years ago | (#19896883)

Remember that OSS is mostly about developers scratching an itch.
Look at the major OSS projects, such as GCC and the Linux kernel. These are not just developers "scratching an itch". On these two projects alone, there are hundreds of full-time OSS developers employed by companies like Red Hat, Intel, Apple, Google, and IBM, as well as by universities and research labs around the world.

Re:Read between the lines people! (1)

jlarocco (851450) | more than 7 years ago | (#19897373)

Look at the major OSS projects, such as GCC and the Linux kernel. These are not just developers "scratching an itch". On these two projects alone, there are hundreds of full-time OSS developers employed by companies like Red Hat, Intel, Apple, Google, and IBM, as well as by universities and research labs around the world.

Yeah, but even GCC and Linux started out as developers scratching an itch. With few exceptions, almost all OSS starts out as developers scratching an itch. Sometimes companies step in when they think they can make money off something, but usually they don't.

Well, Richard had to hawk his turtles too (2, Informative)

tjstork (137384) | more than 7 years ago | (#19896625)

Times like this make me realize that the end result of capitalistic software and open source software is really naught if you are on the losing end. If nobody likes your work, you are not going to be funded, and that's really what seems to be happening here.

The premise for shutting down the project is correct. Multiple cores all but eliminate the need for the most extreme clusters. Throw PCI-X graphics cards into the mix, and you have even that much more computing power. That's not to say that there aren't applications that require clustering, but, those who make those applications probably are going to wind up writing their own distributed processing framework anyway that is tailored to their needs.

Sometimes the turtles are just destined to be soup.

Probably the biggest thing that it needed (2, Informative)

pantherace (165052) | more than 7 years ago | (#19896631)

Was a switch to 2.6. I personally was using it in 2.4, however, when 2.6 rolled out, OpenMosix wasn't going to support it for some time. This caused a lot of users to stop using it, because for NOW, there wasn't really a way to justify staying on 2.4, when the responsiveness of 2.6 was so much better.

I see now that they have an alpha version out for 2.6.

Note that 2.6.0 was released in 2003.

Hi. (0)

Anonymous Coward | more than 7 years ago | (#19896653)

NO WAY is FUNNY. ;) What are you looking at me for?

a little inflammatory (4, Informative)

pavera (320634) | more than 7 years ago | (#19896675)

The article summary was certainly an eye grabber... but, the truth is, I've deployed quite a few linux HA and load balancing clusters. I have also installed a couple openMosix clusters. While it may be sad that openMosix is closing, the vast majority of clustering cannot be handled by openMosix. It is designed as a parallel processing cluster. I would say 99% of clusters are of the HA/load balancing variety. IE, I've got 3 web servers and I want to distribute the load between them. openMosix cannot do this, it isn't designed for it. Or I have 5 DB servers and I want to distribute load/perform replication. again openMosix does not do this. It is a "processing" cluster. IE I have this huge data set, and an application which will split up that data set and do some processing on it. Think SETI@home except, you don't want to send it to people's homes, you just want to run a single process which will send jobs off to other nodes for computing. The only thing I ever successfully used openMosix for was a compile cluster, and for that it was nice, but even for regularly compiling KDE, it wasn't much worth the effort to get the cluster running for the time it saved in compiling.

At the time I used it it couldn't migrate web server processes or db server processes, so it was useless for what I do most of the time.

Re:a little inflammatory (1)

dodobh (65811) | more than 7 years ago | (#19897037)

Clustering can be of multiple types. There is a market for highly parallel processing. Think video processing, weather predictions ...
HA and load balancing is a different type of clustering.

Re:a little inflammatory (4, Interesting)

sl3xd (111641) | more than 7 years ago | (#19897231)

I do highly parallel processing. The industry as a whole has moved in a different direction (which is, oddly enough one of the reasons the project is shutting down). We use MPI, which is one of the things that mosix was supposed to let you avoid. There are other ways to maintain a system than the "single system image." Mosix had problems with performance, which is an effective way to ensure it won't be used in high performance applications.

And it's no fun to develop something you know isn't going to be used, as the supercomputing 'industry' isn't moving in the same direction that Mosix was heading.

Wrong and wrong. (2, Insightful)

Ayanami Rei (621112) | more than 7 years ago | (#19897393)

Video processing is not done through multi-processing with shared memory. It's done in batch, in a grid-type environment.
Weather prediction almost certainly uses special-purpose math libraries (ScaLAPACK, etc.) in a MIMD environment.

Re:a little inflammatory (4, Interesting)

sl3xd (111641) | more than 7 years ago | (#19897181)

The big thing I'd add is that all of the high performance clusters I've seen don't use Mosix (open or otherwise). The reason is that while mosix makes some administration tasks easier, it doesn't address the single most important thing for a HPC cluster: Performance.

The point of mosix is to avoid using a library (such as an MPI implementation) to handle parallel apps, and to make managing a cluster 'easier'.

The problem is that the performance just isn't there, and that the 'industry' as a whole has overall chosen to use MPI to handle parallelism, and use various other methods to manage the cluster.

Bottom line: The industry they targeted didn't move in the direction mosix was headed (which is exactly why the developers are shutting it down).

Important info (dammit, I have modpoints) (3, Insightful)

mr_tenor (310787) | more than 7 years ago | (#19896755)

As usual, people are posting replies without any clue about the actual situation (or at least the claims of important people involved)

See http://mulix.livejournal.com/199931.html [livejournal.com]

"Now the real project can get the credit it deserves. I hate it when people steal credit. It was so annoying to read interviews where it was claimed that behind openMosix are years of research, when all this research was actually behind MOSIX."

Imagine a (5, Funny)

Tablizer (95088) | more than 7 years ago | (#19896801)

and they lack the means or motivation to continue

See what happens when you *stop* imagining a Beowulf Cluster?
   

the danger of OSS (1)

timmarhy (659436) | more than 7 years ago | (#19896813)

This highlights a very real danger with OSS - that the project will collapse under you and you'll be left with 2 choices

1. a painful migration, and i mean painful in terms of giving birth or passing a kidney stone

2. maintain the code yourself, which could be even MORE painful and costly.

yes i know everyone will jump up and down about how this could happen to any project, but folks lets face facts here - In OSS projects where no one is getting paid to write code, you could possibly be hinging a key part of your opperation on the hope that a bunch of geeks don't tire of their little project.

And the "but you can maintain the code yourself" argument is a loser - it's often far more expensive to code yourself then to just pay MS and friends for a solution.

obviously this doesn't apply to huge projects like apache or mysql who have corperate funding.

Re:the danger of OSS (1)

mudshark (19714) | more than 7 years ago | (#19896959)

And EOL doesn't happen in the world of proprietary software.

Right.

Re:the danger of OSS (0)

Anonymous Coward | more than 7 years ago | (#19897123)

OF course it does. But it doesn't happen because the company making the software is bored with it or they just don't feel like doing it anymore. "Most" proprietary software comes with support agreements with guarenteed minimum support lifecycle. eg. just like every single MS product or Redhat product or IBM product. You won't wake up tomorrow with a purchase from any of those companies where they just decided to EOL a product tomorrow because they are bored. This gives companies time to plan migrations or upgrades with some advance warning.

Re:the danger of OSS (1)

timmarhy (659436) | more than 7 years ago | (#19897317)

learn to read. i clearly stated MS EOL'd win98 after 9 years. it was in the first sentence. kind of hard to miss.

Re:the danger of OSS (1)

timmarhy (659436) | more than 7 years ago | (#19897331)

sorry my mistake. i was confusing you with another thread i replied to.

my point is still valid though. MS has a much much better EOL policy then any OSS project.

Re:the danger of OSS (0)

Anonymous Coward | more than 7 years ago | (#19897075)

And the "but you can maintain the code yourself" argument is a loser - it's often far more expensive to code yourself then to just pay MS and friends for a solution.
Really? So what price did Microsoft quote you for MS Mosix 2009 (Data Center Edition)?

Re:the danger of OSS (1)

timmarhy (659436) | more than 7 years ago | (#19897363)

why would MS touch an open source project?

solution = prepackaged software. a little intellgence on your part would be appreciated. .

use xcpu (2, Informative)

warrior_s (881715) | more than 7 years ago | (#19896893)

Guys who used to develop BProc now are focusing their efforts on http://xcpu.org/ [xcpu.org]

It's about the architecture (1)

Chris Snook (872473) | more than 7 years ago | (#19896909)

Look at their versions. The latest release is 2.4.26-based, with a 2.6.15 beta. Clearly it's an invasive patchset, which must be difficult to maintain. Back when the project started, this was probably a worthwhile effort, but the HPC world has changed.

My guess is that the advent of commodity NUMA hardware (Opteron) motivated more development of MPI applications and libraries, since HPC workloads often perform better on NUMA hardware using MPI between the NUMA nodes. If all your applications are MPI-aware, there's a lot less motivation to use an SSI solution that's a couple years behind on hardware support and a pain to maintain.

Use, Buy, D.I.Y, or Get Over It (2, Informative)

cmholm (69081) | more than 7 years ago | (#19897015)

Geezuus, it's not like OpenMOSIX is unusable as is, or that there aren't alternatives. For that matter, while coding one's own cluster controller isn't trivial, it isn't string theory either. Our shop has released (eg. given away) two schedulers, and we've got another that's stayed in house. When I've strolled the booths at the SuperComputing conference, it seems that every other university is giving away their own cluster controller.

OpenMOSIX is neat, but it ain't the end all be all, and it's been my experience that any shop that's serious about running a cluster manages to find/attract someone with the chops to get it up and running. Can just any elementary school pull one together for "free"? Maybe not. For them, there's Pooch [daugerresearch.com] or AppleSeed [ucla.edu] .

Process migration? (1)

Michael Woodhams (112247) | more than 7 years ago | (#19897077)

I've used a 64 node Beowulf cluster [massey.ac.nz] on occasion. The queue was generally full of single-node jobs and multi-node jobs hardly ever ran. There really is no good way of scheduling in this environment without the ability to suspend a job on one node and later resume it on another. So far as I know, our scheduling software was not capable of this. (Fortunately, I was just submitting single node jobs.)

Starting with the information in the summary, I spent a few minutes web searching. "bproc" appears not to be capable of this: it just means that your primary node can "see" the processes running on other nodes, so you can use 'ps', 'kill' etc. on them. However OpenSSI [wikipedia.org] has "process migration". Is this the ability to move processes from one node to another?

Re:Process migration? (1)

r00t (33219) | more than 7 years ago | (#19897419)

Yes, process migration is exactly that.

It's coming to standard mainline unmodified Linux too, judging by comments made by people supplying the container patches that keep getting accepted.

Re:Process migration? (1)

Nuke Bloodaxe (582098) | more than 7 years ago | (#19897445)

Hi, David here, when was that?

When I was working on the cluster during the first 3 months of 2003, and setting up various parts of the software ( a few custom code alterations needed to be made to PBS, as it had a very serious bug that stopped the system from functioning properly at all ), I was sure that there was a bit more than just single node work occurring. You are correct about suspending a node etc, but then again you could also choose a node or node set to run jobs on ( and given the price to run jobs on the cluster it really should not have been too much of an issue unless 16 people had 4 node jobs running at once ). Personally just adding the job to the queue would be sensible, as it would allow the job to run when a free node, or node set, was available.

Check http://helix.massey.ac.nz/pbs.html [massey.ac.nz] As it gives a good overview of how the scheduling system works.

Additionally its a 65 Node system, and has since been superseded by Helix 2, much like the Sisters cluster was.

Naturally this will have changed over the last 4 years, but I'm pretty sure it's not quite as bad as described.

Regards,
David Keenleyside B.Sc. CS & IS

How To Obtain (0, Redundant)

clang_jangle (975789) | more than 7 years ago | (#19897143)

How to obtain that very hard to obtain OS:

http://www.mosix.org/txt_grid.html
http://www.mosix.com.au/prices.html

In other words... (0)

Anonymous Coward | more than 7 years ago | (#19897267)

... The developers have found a new, exciting and more importantly PAID job.

You can congratulate them for their career move at their workplace, the local MacDonalds. You won't get fries for free but nothing beats asking for a Beowulf Cluster of fries.

openMosix != Beowulf (2, Informative)

Anonymous Pundit (161899) | more than 7 years ago | (#19897389)

This item probably shouldn't be tagged with Beowulf.

Most Beowulf clusters run parallel codes written to use the Message_Passing_Interface (MPI) [wikipedia.org] . MPI programs really don't want to be migrated to different nodes while they're running, so load management is achived through schedulers such as Grid Engine [sunsource.net] , TORQUE [clusterresources.com] , and others. These schedulers avoid the need for process migration by preallocating the resources (compute nodes) in advance, and prevent the load imbalance from happening in the first place. openMosix waits for the imbalance to slow down the computation before it migrates a process to relieve the problem.

If you check the archives of the Beowulf mailing list [marc.info] , you'll see that while the Beowulf community knows about openMosix, very few Beowulfers use it [marc.info] .

In summary... (4, Insightful)

Rakishi (759894) | more than 7 years ago | (#19897407)

Let me explain the reason for their decision in a sane way as I see it.

*MOSIX was supposed to provide an EASY way of doing clustered worth. Low over head in terms of coding and administration. It was aimed at MODERATE clusters not massive beasts as it lacked performance/efficiency. While two extra machines may be worth the lower overhead two hundred probably are not so the immense clusters used other methods.

Advanced in computing, multiple cores and so on, have killed this low-to-medium cluster market NOT clustering as a whole.

Yes there are tons of things that still need clustering, think web data for example for a new one, but they are large and even larger. They need performance and so *MOSIX is not what they are looking for.

In other words the market for *MOSIX is effectively dead thus the project is joining it.

Moved on to virtualization? (1)

megabyte405 (608258) | more than 7 years ago | (#19897415)

Am I the only one who noticed that Moshe Bar (the leader of the openMosix project) has gone on to found three (!) virtualization companies, including Xensource? http://en.wikipedia.org/wiki/Moshe_Bar [wikipedia.org] It seems his professional interests have moved on as of quite some time ago, and that this is merely "catching up."

I've looked into attempting an OpenMosix cluster before in my free time but the lack of a 2.6 version made it hard for me to justify the time - as all the work I do from day to day is on a 2.6 system, so I couldn't really gain anything from it. The project has seemed "dormant" for a while. Perhaps this announcement will spur someone on the sidelines to speak up and take the reins, and make openMosix a viable solution for modern (software) systems again: I have a whole pile of spare computers just itching to be run at the same time :D
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>