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.'"

Comments

Hmmmm

[G-funk]

Judging from the photo it seems to be a new form of 3d tetris.... This shall definitely shape the future!

Seen before

[SilentChris]

Uh, you mean like this OS?

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? :)

Re:repeat?

[Hal-9001]

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...

And what about the bandwidth?

[NOT-2-QUICK]

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!

Re:And what about the bandwidth?

[Salamander]

I wonder how upset this individual in Helsinki would be if Mary decided to format her hard disk in the midst of his movie

The Helsinki user is no worse off in this scenario than if Mary's machine were a web server.

I wonder how much bandwidth that cost to prevent this 'just in case' scenario?

We all know that such "just in cases" do actually occur. The only solution to data-loss is redundant copies of the data, maintained either manually (explicit backups) or automatically (transparent mirroring or replication). The authors' idea is to go for automatic replication, and once you have that you might as well use the replicas to improve performance by allowing them to serve data to nearby nodes. This can actually result in less overall bandwidth than traditional approaches, because each node is going somewhere relatively close to get data instead of choking up a central server.

That actually highlights a flaw in the example as given in the article. It would be quite abnormal for someone in Helsinki to be going half-way around the world to get the data, because there should be a nearer replica. It would be more accurate, though perhaps less compelling, to say that Mary's machine was being used as a "staging area" for other local users watching the same movie from Helsinki that Mary just watched ten minutes ago. That would IMO convey the idea of an ISOS (actually the data-store part of it) actually reducing network bandwidth while also improving robustness.

How to Earn that Karma!

[Cheshire Cat]

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.)

Not pratical

[lkaos]

This article makes one fatal assumption: Consumers will always purchase more powerful equipment than they need.

The time of super fast home-PCs is likely to not last very long. The incoming .NET and dotGNU waves are likely to make thin clients much more realistic.

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...

Re:Not pratical

[Jerf]

The time of super fast home-PCs is likely to not last very long. The incoming .NET and dotGNU waves are likely to make thin clients much more realistic.

Can you back this up with any real facts? Today, for $500, you can own a bare-bones Athlon system, which 20 years ago was a supercomputer, minus a bit of memory.

Even after we hit the Fundamental Barrier, whenever that is, computers will continue to improve for a while due to architecture improvements and innovative designs (like 3-D chips, currently totally unnecessary but providing one road for expansion in the future).

It gets to the point where on the consumer level, in a very short period of time (specifically, *before* .NET takes off), it costs very, very little to put in an Athlon, versus a Pentium 100, and that cost is swamped by the display cost. You still need memory on the client side for buffering. You still want a hard drive on the client side for other buffering (like video; a one minute buffer fills RAM pretty fast, but on any conceivable real-wrld future network, we'll need those buffers).

Maybe YOU call a 4GHz Athlon II w/ 512MB of RAM and a 100GB hard drive a thin client, useless to Mary. I call it a dream come true. You have to postulate a Major Breakthrough within the next two-to-three years in display technology for the cost of the display not to swamp the cost of at least (more realistically) a GHz machine with 128 MB of (fast) memory. We'd probably know about it already. So, do you buy the $200 "thin client" that can't do anything on its own, or the $235 "I'd kill for this machine in 1985" that runs fifty times faster, and feels ten times more responsive?

(I made a couple of assumptions in this post. But one way or another, Mary needs a super computer in her home. Either for use that looks like modern use, or to serve as the central server for the house. I, and many others, even amoung the computer non-savvy, will NOT farm my data out to a foriegn entity! .NET does not eliminate the need for fast computers, it just moves it. And the need for more power will be with us for a while yet. We're in a computation bubble here, but voice technology, video streaming, REAL teleconferencing, better video games, and a lot of truly desirable things are still waiting for us over the computation power horizen. And that's just the applications we KNOW about...)

How to Earn that Karma!

[talonyx]

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.)

We don't want "The Network As A Computer"

[refactored]

Good Grief! People's memories are short.

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....

• What I do and can do... (I want unlimited freedom and choice)
• What intellectual product I create... (It costs nothing to make another copy, so why limit distribution?)
• What hard product I create... (It costs much effort to make a copy, and requires hard inputs.)
• What I own... (What I control)
• What is private... (Thoughts and activities that concern me only thank you)

Remove the confusion between the above items and the desire for silly things like "The Network Is The Computer", DMCA etc goes away.

Re:We don't want "The Network As A Computer"

[gwernol]

We don't want "The Network Is The Computer". Remember mainframes? Remember how we joyfully fled from them?

And remember what happened when the Internet came along? Everyone suddenly wanted to be part of a network of machines. Of course the Internet is a diverse set of services running on a diverse and redundant network of machines rather than dumb terminals attached to controlled and homogenous hardware, so its a great step forward from the days of mainframes. Nevertheless the Internet is very much a distributed computer system.

When I use Slashdot I am consuming resources on a remote computer. These days I probably use more CPU power and storage that lives out on the Net than lives on my machine. I don't know about you, but I love it. Much better than the days of standalone machines.

What has happened is we've moved from the days of monolithic, tightly controlled mainframes and terminals, through the personal computer revolution and on to a mixed peer-to-peer and client-server world that gives you the advantages of both approaches.

Of course there are issues, and security and control are amongst the biggest. But these can be solved ultimately, and I no more want to go back to standalone PCs than I want to go back to mainframes.

What we want is to really own our computer power.

Then disconnect your machine from the Net, and you will be happy. However don't presume to speak for the vast majority of computer users who seem extremely happy to be part of a large, distributed network of machines and systems.

Re:There are two types of people out there......

[Radical Rad]

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.