Slashdot Mirror
Scientific American Article: Internet-Spanning OS
Hell O'World writes: "Interesting article on Scientific American outlining what they call an Internet-scale operating system (ISOS). 'The Internet-resource paradigm can increase the bounds of what is possible (such as higher speeds or larger data sets) for some applications, whereas for others it can lower the cost.'"
Judging from the photo it seems to be a new form of 3d tetris.... This shall definitely shape the future!
Send lawyers, guns, and money!
This sounds like (among other things) a larger-scale Seti@Home project - sharing your unused cpu cycles to solve larger problems. I'm not sure how well this would be received, especially given the recent concerns over what these clients are actually transmitting.
Sinepaw.org: Grape Winos
I personally don't like the idea of my OS being spread across multiple machines, or other people being able to use my computing power. If I'm not using my computer, I don't want others using it, reducing it's lifetime. I like knowing that everything I do is controlled by me, on my system. It's a little unnerving to think that my files would be distributed all around the world on other machines. (can we say security?) No thanks, I'll stick with how I'm setup now.
WWJD.... for a Klondike bar?
I mean I get enough junk e-mail as it is, without unethical crooks having an entire OS dedicated to the task...
...Oh, you said spanning OS.
Nevermind.
STOP MISUSING APOSTROPHES, YOU MORONS!!!
Extraordinary parallel data transmission is possible with the Internet resource pool. Consider Mary's movie, being uploaded in fragments from perhaps 200 hosts. Each host may be a PC connected to the Internet by an antiquated 56k modem--far too slow to show a high-quality video--but combined they could deliver 10 megabits a second, better than a cable modem.
I suppose that's great and all, but what if Mary is on a 56k modem? Doesn't really help all that much. I do understand the point they're making though.
WWJD.... for a Klondike bar?
With license v6 by M$, if you install it on your network, and run any other M$ product on that network (even back to Win 3.1), then the license is upgraded to v6 for all of those machines. Where is the boundary? If I do a VPN across the internet to another machine on another LAN, does that mysterious license switch occur? If I am globally connected to many machines on the internet, does the license switch occur on all of these machines?
Kickstart
Wow, 3 years on Slashdot and this is the first time I've caught a duplicate story before anyone else. What do I win? :) A free Kuro5hin.org account? :)
Yes, the exact same article was posted as a /. story here about three weeks ago (under almost the exact same title!) and I could swear it was mentioned in a comment in this story (posted by timothy!), although I can't seem to find that comment right now...
"It take 9 months to bear a child, no matter how many women you assign to the job."
From the supposed real-life example in the article:
"Its disk contains, in addition to Mary's own files, encrypted fragments of thousands of other files. Occasionally one of these fragments is read and transmitted; it's part of a movie that someone is watching in Helsinki."
I wonder how upset this individual in Helsinki would be if Mary decided to format her hard disk in the midst of his movie... Oh, but you say that the same information is distributed on other workstations as a redundancy precaution. I wonder how much bandwidth that cost to prevent this 'just in case' scenario?
While I can certainly appreciate the added value of distributed processing power and multilocational data sources, exactly how is having these massive amounts of data running over the net affecting bandwidth availability?
In my opinion, the lack of a truly distributed ISOS is a bit trivial until we achieve a higher grade of internet connectivity for everyone!
Beer is proof that God loves us and wants us to be happy. -- Benjamin Franklin
about 15 years ago. Down to the gee-whiz! Jetsons prose.
Has Scientific American become nothing but a speculative fiction and PR site for political movements and corporations.
Untill the bandwidth/price ratio available for internet connections grows significantly higher, at present there are only a few exceptional cases where the cost of the data distribution is low enough to make internet distributed computation feasable.
The same applied to clustered storage, with the added problem of the latency to access such storage.
This is not, unfortunately, a tool for helping the average computer consumer. It may, however, be useful for SOME scientific computational problems (ie: ones doing heavy analysis of easily paritionable data), but those are certainly in the minority.
Unfortunately the speed of light over any significant distance soon brings a halt to the scalability of most problems over a widely distributed system, producing a minimum latency which causes the scalability of the system to stop. As computers get faster and storage gets larger this point of decreasing returns gets lower.
Now if we throw in the legal aspects... Can you see the ISP's liking this? how about companies whos equipment is used without their knowledge, and who do we blame for the illegal pr0n being stored unknown to the user on their equipment?
We should not be trying to find ways of consuming bandwith, as it is going to become a more and more valuable resource as computers get more powerfull, instead we should be looking to minimise the bandwidth consumed for given services.
If computers were not still scaling at the rate they are, this may be a useful idea, but that won't happen for some time.
Sorry for going off-topic, but I just have to grieve any time I see anything about my former favorite magazine. Before computers, walking around reading one of these was how you knew who the real geeks were. Where once you had Nobel Prize winning contributors writing articles that took a week to digest, now you have watered down fluff comparable to Discover or Newsweek. Next time you come across an issue printed before 1985, pick it up and learn something.
This quote sounds like it came straight out of an article about linux. The only differance being that linux is not restricted to the limited set of applications it is capable of running.
If linux is struggling (up to this point) to get mass acceptance and use, I can't see an ISOS getting off the ground for a long time yet or ever.
As other posters have pointed out this is a duplicate article. But hey, turn this repeat to your advantage! Go read the previous posting and repost all the +5 posts as your own, then watch the karma roll in! :)
(Yeah, its a little off-topic. I'm sure the mod's will see the funny in it.)
Last night I shot an elephant in my pajamas. How he got in my pajamas I'll never know.
This article makes one fatal assumption: Consumers will always purchase more powerful equipment than they need.
.NET and dotGNU waves are likely to make thin clients much more realistic.
The time of super fast home-PCs is likely to not last very long. The incoming
There is absolutely no reason for 'Mary' to have so much computing power since she doesn't need it. The only real limiting factor today is bandwidth which this article assumes anyway.
What is probably likely in the future though is a more distributed OS. One that is truely network transparent in every facet of operation. I believe there are some rumors floating around about MIT working on something to this effect...
int func(int a);
func((b += 3, b));
I was really worried for a second there, I thought the headline was "Internet spamming OS".
That's one variant of NetBSD we DON'T need developed...
Recursive: Adj. See Recursive.
I will pay for journalistic integrity.
[o]_O
Wouldn't a newly made distributed system either be sued out of existence by people in power, or controlled from its inception by them? I can't picture it working beyond a Distributed.net/SETI kind of thing.
The financial aspect of it is quite interesting though, information and media could be "virtually free" because of your essentially leased out idle computing resources.
As other posters have pointed out this is a duplicate article. But hey, turn this repeat to your advantage! Go read the previous posting and repost all the +5 posts as your own, then watch the karma roll in! :)
(Yeah, its a little off-topic. I'm sure the mod's will see the funny in it.)
Let's notate your Linux box as floodge(0), and ISOS as floodge(1). This higher-order OS would be floodge(2).
It gets better. Now consider an OS of order floodge(N), where N is an unimagineably large but finite number. This would harness the power of millions ** N computers! Truly outrageous horsepower; more teraflops than there are electons in the universe. Just think of how many extra-terrestial intelligences we could discover per second!
Let's just talk about her decoding video and sending it to someone.
Totally uncompressed video is FUCKING HUGE. Basically imagine the size of a bitmap the same resolution + bpp of the video, then multiply that size by 30*seconds of video (for 30 fps video, which is pretty standard I think).
So she could decompress it, and then if she wanted to send it to this Finnish guy there would either have to be a T3 or so between them...
He was probably just watching some porno anyway.
Omnia vestra castrorum habetur nobis.
Heck, as far as simple file-sharing goes, it sounds like Freenet I. (I realize the article is about an operating system, but this discussion seems to be mostly about sharing files) As far as security goes, currently existing file-sharing programs allow you to *choose* which files you want to share, and which you don't, and other users can only request one of the files your computer says it has- no one has actual access to your hard drive*). I would imagine that any decent processor-sharing program would allow similar customization- setting the maximum amount of processor time used, how long the computer would have to be idle before it kicks in, temporarily disabling it, etc. As for wearing out your computer (as someone else mentioned earlier), come on! You'll probably upgrade the old components long before they burn out due to a bit of extra use. Most people run screen-saver programs that keep the processor busy during idle times, anyway. Might as well get some use from it. As for distributed Operating Systems, I agree with whoever said that bandwidth was a more important factor than the program itself. Until everone is connected at 10 Gigabits or so, distributed programs will probably only be used for large, slow things like number-crunching and file downloads - not OS's, which require an immediate response. Something to keep in mind for the future, though. *Barring potential glitches. Really, though, I haven't heard about many security problems in the current programs.
We don't want "The Network Is The Computer". Remember mainframes? Remember how we joyfully fled from them?
What we want is to really own our computer power.
We want a very clear sense of "This is my computer" and "This is my data". I can do what I like with it.
Think folks, what is all the fuss about security and file sharing? Ownership. This is my data to own (keep private) and my data to share (if I choose).
Complexity and installation difficulties steal our sense of ownership. When the computer is a burden, we don't want to own it. Complexity robs us of choice.
The correct fix is not an ISOS, or retreat to mainframe days. The correct fix is to simplify and make things easy.
I don't want my work computer to be my home computer. My employer and I definitely want a strong sense of separation on that front thank you.
Forget these silly pipe dreams, and concentrate on easing the pains of ownership so that we have strength to share.
All this is a silly confusion over....
Remove the confusion between the above items and the desire for silly things like "The Network Is The Computer", DMCA etc goes away.
Actually, instead of defragging, it would just frag it, instantly :-)
available now. windows + code red.
Crud, does this mean I need glasses? I wondered where the piggy icon was.
funny munging
I'm pretty sure that you could make "not for resale" part of the contract.
Devoting compute cycles to specific, worthy causes is great, but the point of an ISOS would be to make all connected hosts more powerful and efficient. If I want to factor a large prime or predict the weather, I might have hundreds or maybe thousands of otherwise idle computers available to help with the task. So each processor is constantly busy.
Privacy is very important but can certainly be worked out. For one thing, data could be stored in "bit stripes" so that each byte of your data is split into 8 separate streams but stored in more than 8 foreign hosts for redundancy and availability reasons. In that way no one could reconstruct any portion of your data from fragments on their drive and no laws could be broken by storing chains of bits.
Also private and public space could be partitioned off so that things you want kept on your system would stay there and only data associated with your weather predicting program would get stored on the ISOS. And quotas would need to be enforced so that if you donate 100GB to the ISOS storage then you may store, say 30GB (due to redundancy) in the distributed system yourself.
And perhaps your CPU's MIPS rating and uptime could be tracked to keep things fair. Then it would be almost like your computer storing up its processor cycles and getting them back all at once when you have a job to run. Grid computing makes sense and a World Wide Grid could make sense if it is feasible and the logistics could be worked out. Imagine everyone everywhere having the power of a supercomputer at their disposal.
Just type find with no arguments and you can see every file on every computer on the net...
There are still no simple ways to use a pair
of computers on the same desk efficiently, why not start there?
The future isn't what it used to be.
People either aren't understanding, or aren't reading properly. All the computers wouldn't be spread out everywhere, only what people choose to put on the net. Your operating system, files, programs, etc. are still on your hard drive, but you can choose to sell extra space on your drive in exchange for some cash, and vice-versa. You can buy a gig of space spread out over the net to store some extra files on, and your files end up in tiny fragments on hundreds or thousands of other computers like yours.
You can buy a gig of space spread out over the net to store some extra files on, and your files end up in tiny fragments on hundreds or thousands of other computers like yours.
And how is that a GOOD thing?
If I need a gig of space, I throw out a gig of crap.
If I am out of crap, I can spend $50 on an extra hard drive. Or $0.20 on a CD-R.
The only way to make distributed storage appealing is to make it so vast that nothing I can reasonably buy will compare with it, and that seems unlikely. And if it DID happen, I'd need a fat pipe to match.
In the end, I want to keep my computer to myself, except for the http server I run.
because to have an effective OS, you would need to have trusted access to those resources....raise your hand if you are going to trust some stupid OS that you have no control over to use your spare Proc cycles, memory space, and Hard drive?
I am the Alpha and the Omega-3
Popular Power had a java-based client. It basically ran off a JDK it helped install on your system, not via the browser. (They ran out of money, dunno what happened to the code.) It would run when your screen saver turned on, which I think makes more sense than asking a user to visit a website.
You're missing the real problem with all these distributed approaches. There aren't many corporate commercial computing jobs that are limited by compute speed. High-end server applications are usually most limited by disk I/O rates, which none of these ISOS approaches effectively address.
ISOS is great for compute-bound problems, OK for network-bound problems, and lousy for diskIO-bound problems, while the application portfolio willing to pay for speedup is overwhelmingly the reverse, except for a few scattered niches.
RPM speeds on disk drives don't improve at Moore's Law rates. The CPU isn't the bottleneck, the database is the bottleneck.
--LP
P.S. Also, writing parallel-efficient applications remains mostly "hard."
The distributed file system thing is exactly what FreeNet already does. However, the key differences between local data and network data, which nobody seems willing to address fully, is what happens when the 'net' runs out of space? Some data gets replicated more than other data--typically by frequency of use--meaning data that's really really important to one person may not be available because too many people are watching Britney Spears movies, and they get replicated more rather than the so-called important data.
Replication of data has tremendous cost: bandwidth, time, and storage space. Its retrieval is also non-trivial. Local data is by far more manageable and secure, so much so that a fully distributed system just doesn't make sense. What does make sense is that people would prefer to carry their data with them.
Consider instead, a bootable business card CD burned with your favorite OS, and a key-sized multi-gig USB memory drive. Constrained to something that will fit in your pocket very comfortably, or even in a normal sized wallet, you can have everything the way you want it, anywhere you go. No need to add the complexity of network distribution at all.
Too often, visionaries put faith in a silver bullet to cure all ails. I prefer simple solutions to solve individual problems effectively.
Any connection between your reality and mine is purely coincidental.
Apparently someone took seriously the suggestion of recycling the highly-moderated posts from the previous ISOS thread. The parent is an exact copy of this post by Ian Clarke on that thread.
BTW, the answer to the (implied) question in Ian's original paper is no. A useful "distributed decentralized data processing system" cannot be built on top of Freenet, or any other storage system that drops data as soon as the herd stops requesting it.
Slashdot - News for Herds. Stuff that Splatters.