Towards an Internet-Scale Operating System

gschoder writes: "Two Berkeley computer scientists (including David P. Anderson of SETI@home) envision an Internet-scale operating system to harness the processing power, networking efficiency, and storage capacity of everyone's computers. Scientific American has their proposal."

Comments

It's been done, and no one uses it

[Frank Sullivan]

Massively distributed operating systems have been around for years... check out Tannenbaum's work on Amoeba. Does anyone use Amoeba? No.

This is two days in a row now that Slashdot has posted articles on the great new idea of distributed operating systems that CS theorists solved and have largely ignored for the last ten years. Besides Amoeba, there was the Connection Machine, VMS clusters, and others.

The fact is, massive distribution is of VERY limited use, and doesn't require OS-level hooks - Napster and distributed.net are both prime examples of useful massive distribution without involving the OS at all.

Re:It's been done, and no one uses it

[slithytove]

Some people use it. For instance i use mosix, which transparently migrates linux processes around.
ive also spent a truly innordinate amount of time thinking about installing amoeba, plan9 and others. the reason i havent is that mosix does alot of what i want in a cluster but i dont have to limit my set of apps to those that come with or i can manage to compile in one of those odd OSes.
But with the OSkit and the growing prevalence of platform independant languages (java, python) i can see a time not too distant when the fireball amoeba distro and the linux single system image projects are competing for the average user.
Or maybe we'll get lucky and a project to put together the best features of plan9, qnx, eros and amoeba will take off with a leader like linus.

Re:The first thing I will do....

[Anonymous Coward]

Sure you will, and as soon as your isos-karma runs out your fork will fail .... read the article and think about it.

You wouldn't do this just like you don't

int main() {

while(1){
walk down street giving $1 to each person you meet
}

return(0);

}

Half a picture

[Salamander]

As happens too often, this proposal concentrates entirely too much on distributed computation, and pretty much ignores the problem of distributed storage. They're quite different problems, each requiring its own solution, even though it's intuitively obvious that any true "Internet Scale Operating System" would have to deal with both.

If you're interested in this "other half of the problem" here are some links:

• Farsite (Microsoft; focus on many nodes, not long distances, but still relevant)
• OceanStore (UC Berkeley)
• CFS (MIT)
• Publius (ATT/NYU)
• Intermezzo

There are many more. The bibliographies for the above will mention many earlier systems, while a quick Google search for these project names will show more recent ones.

Plan 9 / Amoeba

[estoll]

Has anyone heard of Plan 9 or Amoeba? Plan 9 is open source and is developed by Bell Labs (i.e., the same people who introduced Unix). Amoeba was developed by Tannenbaum. These have been around for several years and have not caught on yet. I think the reason is because there is nothing to be gained by the home user. Why would someone want people around the world using their computer when they were away? Just thinking about the security risks alone would make me skeptical.

Re:How about desk-sized?

[Holesome]

They make these wonderful contraptions known as KVMs...

I don't think that was what the poster had in mind. Your KVM switch doesn't provide any value other than saving desk space. The article talked about the benefits of redundancy, increases processing power and increased bandwidth.

I imagine what the poster was talking about was having one operating system that would use both computers if they were available but having a complete working system if one was unavilable. So for instance you could power up the second computer with an additional 56K modem and get dual PPP connection without any effort.

Re:Scary...

[ddstreet]

Nope. Cause some l33t h4x0r will have own3d her already.

Micro$oft Press Release #10520

We are happy to announce the immediate availablity of our new distributed computing service! For a low fee, you can harness the power of EVERY computer installed with Windoze XP in the world! Yes, that's right, all their base are belong to us, and you can buy CPU time on 'em!

What's scary is that (except for renting out time) the above is TRUE. M$ does 0wn all Windoze XP systems. And people PAY them for it!!! Inconceivable!

Already happening.

[Tenebrious1]

A couple years ago, a friend sent me a link to a distributed computing (DC) website for cancer research (IIRC). When I looked at the fine print, the DC company was a for-profit service. The cancer research, non-profit, couldn't afford and did not have the technology to run its own DC setup, so signed on with the DC service. The fine print said that 1/5th of the work packets would be for the cancer research, while 4/5ths would be for "paying" customers, who subsidized the other 1/5th share. It did not say who the paying customers were.

After thinking about it, I decided against it. I had no idea who was paying for the other 4 work packets- big tobacco, Iraqi agents doing bio weapons research, Chinese nuclear weapons development. If they had said right out who it was for, I might have still signed up, I really didn't like the way I had to poke through the fine print to figure this out.