Google and IBM to Provide Cloud Computing to Students

John "butter/oreo" Bajana-Bacall writes to tell us that IBM and Google have decided to team up to provide cloud computing resources to participating college students. "Most of the innovation in cloud computing has been led by corporations, but industry executives and computer scientists say a shortage of skills and talent could limit future growth. 'We in academia and the government labs have not kept up with the times,' said Randal E. Bryant, dean of the computer science school at Carnegie Mellon University. 'Universities really need to get on board.' Six universities will be involved in the initiative. They are Carnegie Mellon, Massachusetts Institute of Technology, Stanford University, the University of California, Berkeley, the University of Maryland and the University of Washington."

Comments

Students Response

[Anonymous Coward]

Students to ask what the hell that means.

Many busy contemplating brand new concept: 'clouds of porn.'

Hmmm....

[Anonymous Coward]

John "butter/oreo" Bajana-Bacall

The guy's name, with stupid nickname, is longer than the entire contents of his site.

Cloud computing?

[Anonymous Coward]

Is that web 2.0 compliant?

Re:Cloud computing?

[Nite_Hawk]

From wikipedia:

Cloud computing is a term used to describe applications that were developed to be rich internet applications. In the cloud computing paradigm software that is traditionally installed on personal computers is shifted or extended to be accessible via the internet. These "cloud applications" or "cloud apps" utilize massive data centers and powerful servers that host web applications and web services. They can be accessed by anyone with a suitable internet connection and a standard browser. http://en.wikipedia.org/wiki/Cloud_computing

Re:Cloud computing?

[Anonymous Coward]

It's a very pie-in-the-sky concept, minus the pie....

Re:Cloud computing?

[Anonymous Coward]

How is this different from the concept of SOA? "...software that is traditionally installed on personal computers is shifted or extended to be accessible via the internet". I was under the impression the idea of Grid processing was more about sharing computing tasks...

Re:Cloud computing?

[morgan_greywolf]

Examples: Google Maps, Google Docs & Spreadsheets, etc., and, to greater or lesser extent, Google Earth. All rely on massive computing power connected via the Internet and HTTP.

IOW, didn't the students already have cloud computing? Or is this an implementation of the server side?

Re:Cloud computing?

[jlarocco]

Does that really need a new buzzword? Sounds like the same old shit that people have been doing with the internet for 10 years now. At the very least, isn't that basically the definition of "Web 2.0"? What's the difference?

I was a little worried I had completely missed out on some new phenomenon, but that Wikipedia page has only been around since March. Sounds to me like Google and IBM just want to inspire "OMG!!1 We're missing out on 'cloud computing'!1" in idiot PHBs and investors.

Re:Cloud computing?

[Anonymous Coward]

cloud computing!? isn't this just another name of grid computing, or distributed computing, or clustered computing, or *&^%$#@!?

Re:Cloud computing?

[Kadin2048]

How is this different from the concept of SOA? "...software that is traditionally installed on personal computers is shifted or extended to be accessible via the internet". I was under the impression the idea of Grid processing was more about sharing computing tasks... I don't think there's really a difference. "Cloud computing" strikes me as basically a sexed-up name for SOA, directed towards end-users rather than corporate implementors.

That and SOA is pretty vague. I don't know where you're getting your particular definition from, but there are lots of competing ones; you can include a lot under the umbrella of "service" within "service-oriented architecture." It doesn't refer just to web services or even just to software and IT.

Re:Cloud computing?

[kestasjk]

I think it's because the internet being represented as a cloud in most diagrams, so it means "internet computing". I think "internet computing" might be better, but then someone would decide it takes too long or isn't cool to write "internet computing" and would use "IC".

The best thing to do when someone says a stupid acronym is do just what the GP* did, and ask "Exactly what does that mean?", and don't use it yourself.

* Ahh, the irony..

Re:Cloud computing?

[Dan Yocum]

yes.

I refuse to answer this...

[monkeyboythom]

on the grounds that the Rolling Stones will sue me and everyone else for the use of the word, cloud.

Re:I refuse to answer this...

[morgan_greywolf]

on the grounds that the Rolling Stones will sue me and everyone else for the use of the word, cloud. Pfffft! Get offa my cloud!

What if & modern computing

[Dareth]

I often wonder what form modern computing would be in today if the personal computer had not been so wide accepted. Look around you at the walls. Some of the things you see are very ubiquitous. People take electrical outlets and phone jacks for granted. It is just part of the infrastructure we are used to. Now imagine a computer port next to all the rest. All you need is simple input(keyboard,mouse) and simple output(monitor,printer) devices attached to an adapter that plugs into this outlet. That is all you would need to know about computing. Computing power would be offered by a "Computer Utility" company. They would handle all the technical details. You simply pay your bill and the "technical goodness" comes down the line.

Sure, you certainly pay thru the nose for your time slices of CPU power. But to those of us fortunate to be "Computer Wizards" who live and work at the Computer Utility, life would be grand!

Re:What if & modern computing

[pohl]

In that "grand" alternate universe Lily Tomlin would have done a different character, perhaps, and we'd be accepting abuse from the computing utility company.

Re:What if & modern computing

[Anonymous Coward]

But to those of us fortunate to be "Computer Wizards" who live and work at the Computer Utility, life would be grand!

And here you have identified what I believe is the root cause of the Microsoft hatred around here. The "Computer Utility" is not really any different than the "Glass Computer Rooms" of the 70s and early 80s before the PC came along. So now many of those "Computer Wizards" who formerly inhabited the glass rooms are dyed-in-the-wool Microsoft bashers largely because of Microsoft's role in providing computing power "for the rest of us" (to borrow an Apple phrase).

Re:What if & modern computing

[Anonymous Coward]

You're doing it wrong.

20902723

Re:What if & modern computing

[Animats]

often wonder what form modern computing would be in today if the personal computer had not been so wide accepted. Look around you at the walls. Some of the things you see are very ubiquitous. People take electrical outlets and phone jacks for granted. It is just part of the infrastructure we are used to. Now imagine a computer port next to all the rest. All you need is simple input(keyboard,mouse) and simple output(monitor,printer) devices attached to an adapter that plugs into this outlet. That is all you would need to know about computing. Computing power would be offered by a "Computer Utility" company. They would handle all the technical details. You simply pay your bill and the "technical goodness" comes down the line.

That was Control Data's vision of computing, circa 1967. One supercomputer per city. There's a paper on this in, I think, AFIPS FJCC for 1967. General Electric Time Sharing Services was providing something like that by 1968. The whole "time sharing" business tanked by the end of the 1970s.

Re:What if & modern computing

[Anonymous Coward]

You don't have to wonder, actually. Check out Minitel.

Re:What if & modern computing

[rtb61]

In this case access to cloud computing and you could be paying through the nose for it. Google is a strong supporter of patent first, not invent first legislation, with google's history I would stay will clear of running any possibly patentable work an a computer system that they had anything to do with.

Whilst IBM certainly has a good reputation, you would think they would hesitate to associate themselves with google's privacy invasive history. As for those famous research universities, did they stop to think about the changes to patent legislation, and protecting their student and staff members ideas.

Re:What if & modern computing

[replicant108]

Computing power would be offered by a "Computer Utility" company. They would handle all the technical details. You simply pay your bill and the "technical goodness" comes down the line...

... as you wave goodbye to your freedom.

My experience with cloud computing

[Anonymous Coward]

I got very very hungry afterwards.

Re:My experience with cloud computing

[suitepotato]

Dude, wait... what???

i keed, i keed

[User 956]

industry executives and computer scientists say a shortage of skills and talent could limit future growth

That doesn't seem to have stopped Microsoft.

Re:i keed, i keed

[umghhh]

By definition industry needs something different from what universities provide. That is so because universities teach about things that do exist already not about things that are being developed now (some of them do research too but that is not teaching). The role of a school was always to provide the base on which one can build. The role of a company in it is to provide means to build and such foundations.
Instead of complaining they should use talent and skills that are at place.

I thought that is what the companies normally do. Maybe if the common practice in the industry were not short term profit and black mailing with skilled labour in cheap countries there would be more young people willing to do something else except managing portfolios and takeovers,suing other people and making silicon implants.

Re:i keed, i keed

[jd]

Last I heard, Universities taught the underlying theories and underlying principles, the science behind the practices and the maths behind the science. At least, that is what happened at the University I went to. And, as Inmos mostly worked with recent graduates, I feel confident in saying that many of the better European Universities work this way.

Knowing what was common practice at the time the text books were written is useless. Books take years to write, by people who aren't usually researching at the same time, so must be drawing on older knowledge still. It can then take years more before those books begin to be picked up by Universities. Any book talking about an implementation that existed at the time of writing is talking about an implementation that probably no longer exists at the time of the course.

Theory, on the other hand, changes much more slowly. People knew of electron tunneling long before it started showing up in silicon or germanium arsonide chips. People knew the theories behind good software and hardware designs long before such designs became practical to build. (In the case of software, formal methods are still regarded as impractical in their ideal form, but it is in their ideal form that the best software will be written.) Relational database theory existed long before there existed computers nearly powerful enough to run relational databases.

The theory is the foundation on which all practical achievements can take place. Without a foundation, no stable achievement is possible. That is why a certain range of products made by a well-known company, without regard to foundational knowledge, are of such poor quality. You can't build indefinitely on sand and expect the structure to hold up.

There is no value in teaching "practical" skills at University, because what is practical will have long-since moved on and vanished into the horizon. What you need to teach is how to distinguish what is practical from what is not, how to learn what is practical today (and tomorrow), and how to develop new practices from what exists. Students should be concerned only with learning how to move forward efficiently and effectively, given whatever starting point they discover themselves as having.

But...

[Chris Mattern]

...I wanted Squall computing!

Chris Mattern

Re:But...

[KudyardRipling]

$ telnet d4thc.cloud.computer.lan
Trying dead:beef:deca:fbad:ba5e:ba11 ...
Connected to dead:beef:deca:fbad:ba5e:ba11
Escape character is '^]'
user: root
password: ********
Welcome the the Cyclone Beowulf Cluster. Please don't blow up the world.
#
# hack whitehouse.gov | find football.txt | crack | rsh pushthebutton.sac.mil
Operation successful. You may kiss your ass goodbye.
#

Re:But...

[Aetuneo]

I would rather have Tempest, personally. You can get more power from it (although, admittedly, it tends to overload whatever I/O device you use, and capsize^H^H^H^H^H^H^Hcrash your computer).

John "mayo/chip" Banana-Recall

[xxxJonBoyxxx]

John "butter/oreo" Bajana-Bacall

Are you kidding me? Is this some kind of inside joke or is this guy's name really that messed up?

Open to the public

[proidiot]

While I'm glad they're opening this up to top universities before businesses, I would think that both companies should probably open this up to open source development as well. While dealing with real hardware is ultimately a must for any serious package, it would be nice to have a way to get a package off the ground without killing local resources (not to mention potential advantages in version control).

Re:Open to the public

[Wesley Felter]

Hadoop is already free and EC2 isn't that expensive.

Cloud computing?

[jlarocco]

Sorry guys, I missed the memo. WTF is cloud computing?

A Little Confused

[MrCrassic]

I am confused about the concept of cloud computing. Is it supposed to be similar to that of the famed beowulf cluster, as in making a supercomputing platform out of regular computer networks? Or does it use more powerful computers and cluster them together?

Furthermore, what would be the point of doing this exactly?

Re:A Little Confused

[Prysorra]

It's basically an on-demand ad-hoc supercomputing network.

Almost like clouds appearing where "needed" most according to meteorological parameters.

Re:A Little Confused

[mgblst]

So how is it different to Grid Computing?

Re:A Little Confused

[Duncan3]

It's not. They want you to come back to the central mainframe too.

Re:A Little Confused

[STrinity]

So how is it different to Grid Computing?

To use it, you have to put a Care Bear sticker on your PC.

Terminology? Marketing?

[villy]

This sounds an awful lot like what grids do (and have been doing for a while). I did rtfa and I didn't see much defining "cloud computing" other than "large data centers that students can tap into over the Internet to program and research remotely". Is "Cloud" the new "Grid"?

Now is that the cloud that has the...

[korbin_dallas]

letters 'PFM occurs here' in it ???

That was Thomas Watson Jr's vision.

[mosel-saar-ruwer]

Originally the IBM machines were strictly lease-only [little money upfront, big money down the road].

Then sometime later they moved to the sales model [big money upfront, but little money down the road], and Thomas Watson Jr always felt that that was a disastrous mistake.

In fact, the entire industry [M$FT, Oracle, IBM, Sun, HPQ, Unisys, Google, pretty much everybody] has been working desperately for the last ten or fifteen years to get away from the sales model, and back into the rental/services model - everyone seems to agree that that's where the big $$$s lie.

Re:That was Thomas Watson Jr's vision.

[rayzat]

IBM didn't decide to switch from a lease model to a sales model they were forced to switch to it as a result of the anti-trust lawsuit that was filed against IBM.

Hangon...

[CRX588]

Does this "Cloud computing" require "Tags" in anyway? I mean were talking "Web 2.0" right, someone did mention this in the "Blogosphere"?!

Either way, as long as this stuff does not run on a "Hypervisor" I don't want anything to do with it!

Re:Hangon...

[DrVomact]

Sir, I believe you are trivializing this breakthrough concept. In actuality, "cloud computing" involves a direct high-bandwith fiber optic connection to Cloud-Cuckoo Land. Perhaps the greatest benefit of this new concept is that it supports greatly simplified Power Point presentations—all you need is a really fuzzy picture of just about anything. Then you sketch a little stick figure labeled "user" and draw arrows connecting him to the fuzz. Finally, you add a single bullet item: "Profit".

Meta question

[mosel-saar-ruwer]

Presumably these clusters are for really hard problems - folding proteins, or simulating nuke explosions, or searching for exotic primes, or classifying Lie Groups, or proving Four Color theorems, or whatever - i.e. presumably these programs are expected to run for a long, long time before they terminate.

On the other hand, a fellow named Alan Turing once proved that we can't know whether an arbitrary program will ever terminate.

Now here's the question: If you allow a student onto one of these clusters, and if his program keeps running and running and running and running, with no apparent end in sight, then how do you know whether there's actually an infinite loop within his program, or whether it's just a very, very, very hard problem he's trying to model?

So if you are one of the lucky few who gets chosen [or at least pre-selected] for this sort of thing, then will you have to submit a "proof" of the finiteness of your program before you're given the green light?

And will they provide any formal "template" within which the student could "prove" finiteness, or at least offer an outline of [a hope for] a proof?

And might there be some set of "mileposts" which the program is required to meet in a given time, and if, as it runs, the program fails to meet a milepost in time, then it's given the heave ho?

In a similar vein, are the lucky few required to "prove" that they have used all of the fastest known algorithms for each of their calculations?

Just as an example, have you ever timed the computation of "n choose k" using the actual multiplication & division of the factorials, and then compared it to the speed of Pascal's triangle?

Or tried anything in signal analysis without the benefit of O(nlog(n)) algorithms?

Re:Meta question

[deander2]

Or tried anything in signal analysis without the benefit of O(nlog(n)) algorithms?

  O(n*lg(n))

nitpick, i know, but "log" w/o specifying the base usually means base 10, and i assume you mean to say base 2 (which is usually written "lg")

Re:Meta question

[g0at]

What does "meta question" in your subject line mean?

-b

Re:Meta question

[Angostura]

Easy. Just ask the student for their graduation date, If the app hasn't reported by then, either terminate or send an e-mail asking if the results are going to form part of their post graduate studies.

Re:Meta question

[Ckwop]

So if you are one of the lucky few who gets chosen [or at least pre-selected] for this sort of thing, then will you have to submit a "proof" of the finiteness of your program before you're given the green light?

The halting problem is actually tractable for the vast majority of algorithms. If you were to select a program at random from the vast sea of possible programs, the vast majority are known to halt, however, the precise percentage is itself in-computable.

Most commonly used algorithms such as quicksort or the myriad network traversing algorithms have a strict proof of correctness. These proofs must include a proof that algorithm terminates on any given finite input, since it would be a bug if the program failed on a particular input!

In fact, modern static analysis tools have pretty much solved the halting problem from a practical stand-point. That said, there are very simple programs whose halting status is unknown. For example, consider:

def f(n):
print n
if n == 1:
return

if n % 2 == 0:
f(n/2)
return
if n % 2 == 1:
f(3*n + 1)
return

Slashdot mangeled the proper indentation but you get the drift. Nobody currently knows whether this halts on all inputs.

Of course, you probably wouldn't do any of this on a real machine. You'd probably pay for time up-front and the program is killed after that time has been exhausted. You probably wouldn't do any static analysis on it.

Simon

Re:Meta question

[Daniel Dvorkin]

It's irrelevant, since lg(n) is a constant multiple of log(n) -- more generally, a logarithm of any base is a constant multiple of a logarithm of any other base -- therefore O(lg(n)) = O(log(n)).

Re:Meta question

[pikine]

Is there a name to this function?

Re:Meta question

[rubycodez]

your reply looked like it would never end

Re:Meta question

[Krakhan]

I thought the issue of the halting problem wasn't for a *particular* program of whether it could be halt, but rather the problem was to come up with a single, general algorithm (or program if you want) that would decide for *all* programs on a given input whether it would halt or not. That is an undecidable problem.

However, although it is possible to prove certain algorithms terminate, but the proof is special for each one that you have to come up with.

With that in mind, I always found the interesting thing the undecidability of the halting problem proves is that it is impossible to show that an algorithm exists for proving whether any arbitrary mathematical statement is true or false. This doesn't mean mathematics in undecidable of course, for the same reason I mentioned above (special proofs for each statement).

Re:Terminology? Marketing?

[heli_flyer]

Cloud computing is the new name for mainframe computing. It's a marketating word, devised by some marketating person.

Google and IBM recruit top scientists for FREE!

[opypod]

What a great way for IBM and google to get the best and brightest to solve all kinds of problems for FREE! Well, it costs them $30 million a year, but that's probably nothing compared to having their OWN employees develop something. Industry has been raping and pillaging academia for years (e.g. the Pharma industry) so why not extend that opportunity to the computer world. GENIUS! I'll have to hand it to IB-Goo (sick forshadowing???), they sure are generous donating all the that computing power (which no doubt they own and control). You wouldn't want to let those researchers have full control of a thing like this ... they might do something REALLY good with it and then IB-Goo will NEVER make any money off those good ideas. I'm not sure how much $$$ Stanford gets for the google IP developed there, but of course, that was back in the day BEFORE "cloud" computing was invented. Such primitive times those days were ... such primitive times.... How did we ever live before the heady days of "cloud" computing.

Re:Google and IBM recruit top scientists for FREE!

[Anonymous Coward]

google has been raping and pillaging open source for years, this is old hat to them...

not once have they released back into the community any portion of the software that they
have modified and that is mission critical to them.

summer of code my ass, that's just window dressing

Re:Google and IBM recruit top scientists for FREE!

[Tyebald]

> I'll have to hand it to IB-Goo (sick forshadowing???) Don't you mean, "GooBM"?

Here ya go

[Anonymous Coward]

Easy computing as a "utility service", 100 buck decent looking machine, preloaded linux, then so much a month for hassle free maintenance and some online storage for your data (cheap price on the hardware with signing up for the maintenance), Designed for computer noobs or folks who don't want *any* hassle. zonbu. This is the grandma/your folks, etc answer, without you having to ever be administrator. Fire and forget.

Just aware of it, I have no ties to them at all

nyuk nyuk

[Anonymous Coward]

teh funny, +1

Have you ever wondered why?

[Anonymous Coward]

We build all those castles in the sky?

Its to be closer to the cloud people. The cloud people are very smart... They invented SOAP, AJAX, Web 2.x, 3.x and 4.x, the flux capacitor and exploding laptop batteries.

All of which the world is not?????? better off without.

Free is hard to beat... but I have to wonder

[NotQuiteReal]

How much to rent a bot net to do "real work"?

Hey, results are results, right? And if it lessens my spam, oh well.

So what is this really about?

[Anonymous Coward]

First I thought they were giving their unused processor cycles away for research. But this doesn't seem to be about protein origami for healing Aids and stuff. FTA it's more like they want those students to play at being Google in their little sandboxes:

This new kind of data-intensive supercomputing often involves scouring the Web and other data sources in seconds or minutes for patterns and insights.

If Google is actually giving them access to their databases, watch out for some juicy torrents.