Slashdot Mirror
IBM Wants CPU Time To Be A Metered Utility
kwertii writes "IBM CEO Samuel J. Palmisano announced a sweeping new business strategy yesterday, pledging $10,000,000,000 towards redefining computing as a metered utility. Corporate customers would have on-demand access to supercomputer-level resources, and would pay only for time actually used. The $10 billion is slated for acquisitions and research to put the supporting infrastructure in place. Will this model revolutionize the way companies compute, or is this plan doomed to be another PCjr?"
This won't revolutionize anything... I remember this when it was called timesharing on mainframes. The revolution was moving away from that model...
In the mainframe world cpu cycles are already a potential billable transaction..
So the concept is old and crusty..
---- Booth was a patriot ----
for companies and institutions that use a lot of heavy computation.
It takes a lot of time, space and know-how to own and maintain big-@ss computers. With broadband connections being commonplace, you could run your own progam remotely, and let a specialist (like IBM) handle all that stuff. And of course, there is value unlocked by having multiple users share common resources.
Of course, the vast majority of companies and institutions (not to mention individuals) use their machines mostly for word processing and surfing the net - and thus they will have little use for this kind of service.
Tor
The concept IBM is going for is to treat IT as another utility. Instead of some small company having to keep an expensive IT staff and maintain their own computers/network/storage, IBM says that it will do this for you. IBM will essentially replace the IT department and let some organization concentrate on running their own business.
The cost saving of such a model (if successful) are quite substancial and will save everyone money in the end.
I think IBM is on the right track with this and they are the only company really positioned to do so.
Let's say IBM is able to set up a way to do what they propose, here's some basic utility concepts I'm curious how they will adddress:
1. Transferring product from generator (IBM supercomputer) to location. If you've just used 1 month of supercomputer time to model DNA folding, how will IBM transfer that data back to you? What if the computations and use are faster than the transmission rate? [Modem vs. DSL vs. T1 line]
2. Dependency - you rely up on natural gas and electricity to be there, and yes they go down, but can they guarentee their utility won't have worse problems - especially if its Windows run and goes down once a week, cutting into your bought utility time.
3. Regulation. Most utilities are regulated, and those that were deregulated have not always worked out for the consumer. Let's say company A gets rid of its expensive infrastructure for computing resources and uses IBM's utility. What if IBM becomes the only utility and charges way more than it should - there's no competition so Company A can't shop around. Along this same vien, if Company A is smart enough, they'll never enter into a utility agreement with IBM if they can generate their own computing cycles and be sure that they'll always be there, versus putting all their eggs in one basket.
IBM's idea may have merit, but anytime someone throws out the idea of a new Utility, that suggests that the resource they're selling is mainstream and essential, and therefore, is treated as a commodity. Those commodities are regulated and made reliable so that they never go down. I can't see supercomputing cycles as being something that is commodity, or for that matter, something I (or any company) needs to buy on a metered basis.
-When going for broke, go for Ithaca!
This is not just about paying the meter. It is about utilizing all the wasted CPU cycles.
No matter how many articles I've read, it always amazes me how few Slashdotters read the article before they feel compelled to post their (usually misguided) opinion. I'm sure plenty do, but there sure are a lot who don't.
IBM is working on the commercialization of Distributed Computing (henceforth, DC). This effort has been around for a while (in a related area, called Grid Computing, which some people use interchangably with DC) in the form of the Globus project, amongst others.
The concept behind DC is essentially a next-gen timeshare-- a distributed timeshare with an abstration layer, if you will. Unlike traditional timeshare, you don't specify where your processing will occur. Unlike existing projects (like folding@home, dsitributed.net), DC doesn't require that you have a parallel, segmentable computing problem.
Let's say (in your best Police Squad voice) I'm a mechanical engineer who's designing a car engine with a few thousand parts. I want to run some simulations on my model to inspect heat flows, vibration, whatever. Car companies (or the little guy with a copy of Catilla and a great idea) don't necessarily have dedicated computing resources to run my simulation. So, until now, I had to band together with a bunch of other mechanical engineers with jobs similar to mine and try to justify a giant simulation node. Or, I might convince management to outsource the computation, requiring a bunch of red tape, NDAs, contracts, negotiation, etc.
Now consider IBM, one of the largest commercial web hosts. IBM maintains giant server farms to support these services. Consider the amount of excess processing capacity sitting in these server farms because (a) a lot of servers are spitting out static pages and (b) extra capacity necessary to cover peak loading for special events.
Expand this idea to include thousands of people who need computation power for discrete, isolated projects and thousands of companies with excess computational capacity. The consumers don't care precisely where or when their computations get completed, they only care that they get done in a "reasonable" amount of time. An intermediary, which it looks like IBM wants to be, can accept jobs from them, break them into as many pieces as they can, farm them out to whichever of their suppliers has excess capacity at any particular moment, combine the results, and return them to the customer.
Even more, IBM can charge more if you want a high priority on your computation or if your job is not symmetric and must be run on fewer nodes.
Actually, if you think about it, IBM is hurting their server sales by advancing this project. Right now, they sell a lot of excess capacity to companies to cover their peak loading. If companies can dynamically purchase exactly the amount of processing they need, that's money IBM's leaving on the table. Now, companies with high-availabity requirements will still purchase their own systems with enough extra capacity to cover their own needs. But, when they're not using that capacity, they'll sell it.
I think IBM saw that the train was leaving the station. They know this technology is coming. And they see that the chance to be the intermediary in this market is worth more than the money they'll lose in hardware sales. And, they know if they don't, someone else will.