Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Cloud Security Supercomputing Hardware Science

'Blind' Quantum Computing Proposed For the Cloud 89

judgecorp writes "Researchers at Vienna's Quantum Science and Technology Center have proposed that 'blind' quantum computing could be carried out securely in the cloud. When (if?) quantum computers are developed, they will be very fast, but not everyone will have them. Blind quantum computing will be useful, because it shows that users can encode 'qubits' and send them to a shared quantum computer to be worked on — without the quantum computer having any knowledge of what the data is (abstract). The data also cannot be decoded form the qubit while it is in transit. It's good to know that quantum computers will be secure when they exist. At the moment, of course, they are even more secure, by virtue of their non-existence."
This discussion has been archived. No new comments can be posted.

'Blind' Quantum Computing Proposed For the Cloud

Comments Filter:
  • quantum hype (Score:4, Insightful)

    by Anonymous Coward on Saturday January 21, 2012 @06:15PM (#38777189)

    Quantum computing is just a rather basic branch of computer science which seems to be winning all the hype in the world at the moment because there's not much sexy in terms of hard non-biological research with a practical slant.

    I'm not quite sure how the output remains unknown to the computer. I know most cloud services either haven't or won't last long enough to give you all the output you were expecting to get for free/absurdly low cost, but I'd be very impressed with a computing system which is able to deliver you something without knowing it's delivered it.

    Of course, there are various computations which can be performed partly by a separate processor without the initial input and final output being known by that separate processor, but there's nothing quantum-y about that.

    • Re:quantum hype (Score:5, Informative)

      by Anonymous Coward on Saturday January 21, 2012 @07:21PM (#38777521)

      As a physicist working in the field, let me correct two things:
      Quantum computing is not a branch of computer science; it is not logic and mathematics. It is a branch of physics. It is also why you find it hard to believe that both the input and output is unknown to the quantum computer.

      @OP: They do exists, just not in stores yet, a natural result of having unsatisfactory shape.. Try paying a visit to a QC laboratory.

      • Re: (Score:2, Interesting)

        by Anonymous Coward

        As a mathematician not working in the field, let me point out that any AC saying "as a X working in the field" probably deserves to have the rest of his post ignored.

        Anyway, the production of quantum algorithms is a routine computer science problem. And the building of a quantum computer is an engineering problem. I'm not really sure what great contribution physics per se is making to quantum computing, i.e. what new science is being discovered and applied to quantum computers, but perhaps you'll enlighten

        • by Anonymous Coward

          I know that the output is known because otherwise it couldn't be converted to classical form for sending across the 'net to the customer.

          I'm sorry but that's nonsense. Even in classical computer science, there are algorithms whose output isn't known to be correct, or even known to exist altogether. This is the norm with so called randomized and stochastic algorithms. They'll give you an output all right, but there's a small chance that the output is meaningless. Or they'll give you an output with a terminating symbol (which tells you that the algorithm actually completed and everything that follows is filler/garbage), but there's a small ch

          • by Anonymous Coward

            This is not a matter of the output being correct or not. Quantum computer doesn't yield an output at all. It yields a quantum state. The measurement performed by receiving end is the output, which doesn't exists before the measurement.

            > However, you are right that one doesn't have to know physics to devise quantum algorithms. One just needs to know and apply the rules, which are sufficiently well axiomatized by now.
            Those rules are called physics.

        • I know that the output is known because otherwise it couldn't be converted to classical form for sending across the 'net to the customer.

          I haven't read the article, but I guess it's not classical bits sent over the Internet, but qubits sent to the customer via fiber (just like in quantum cryptography). Standard telecommunication fibers have been shown to be able to transport qubits reliably enough for quantum cryptography, therefore they should also be reliable enough to transport the results of quantum com

      • by Anonymous Coward

        Quantum computing is not a branch of computer science; it is not logic and mathematics. It is a branch of physics.

        Why the Quantum's physicians didn't name them Quantum Physics, or Quantum Mechanics, or Quantum Chemistry, or Quantum Particle instead of Quantum Computing, or Quantum Logic, or Quantum Math, or Quantum Satisfiability, etc.?

        The minimal quantum unit of information is the q-bit. Is the q-bit a particle in the Physics? I don't think so, the q-bit is in the "Ficticious World" of the Computer Scien

      • Well, "they do exist" is only true if you already call a system of a few qubits, not capable of anything more complex than factorizing 15, already a computer. I'd say we have a proof of concept, but not yet a real quantum computer. Making qubit implementations which scale up to a larger number of qubits is still an active field of research.

    • by m50d ( 797211 )
      One of the interesting things about quantum computing is that it's fundamentally impossible to copy a qbit (hmm, wait until hollywood hears about that). So the cloud service really couldn't have logged what it sent back to you before it sent it.

      More to the point, I suspect this proposal works by sending entangled qbits into the service, keeping the corresponding pairs, and getting back something that can only be turned into your answer by combining it with what you kept. This isn't the same as what you're t

      • Re:quantum hype (Score:4, Interesting)

        by marcosdumay ( 620877 ) <marcosdumay@gm a i l . com> on Saturday January 21, 2012 @09:21PM (#38778125) Homepage Journal

        There are some keywords at TFA that give a hint. The computer is "measurement based", what I'll understand as "the computer only does measurements", also, "without knowing the original states, nobody can decode the output".

        Turns out that are infinite ways (normaly over a finite continuum space) to encode your original bits, and if your computer only does measurements, the answer will be encoded the same way you encoded the data. If the computer operators don't know your encoding, they won't be able to read your data.

        The hard thing is getting those phothons already encoded through the world into the computer, and getting the results back. Also, the above assumes that you can't discover the encoding, but it doesn't survive known plaintext attacks.

    • by jlugert ( 698711 )

      Quantum computing is just a rather basic branch of computer science which seems to be winning all the hype in the world at the moment because there's not much sexy in terms of hard non-biological research with a practical slant.

      Not even wrong. Quantum computation is a parallel process which scales exponentially with the number of computing elements (wetware appears to have this property BTW, speaking of computational biology). Computers can't compete in this space--not with a zillion cores. Quantum states inherently sort out their threads. In other words, it is the next DSP/FPGA/GPU but entirely different--dreadlocks not deadlocks. Had we pursued this path in the 40's instead of computers, we wouldn't be in the jam we are now.

      Fun

  • If we were doing in this in a pre-quantum-minicomputer era wouldn't it be easier to just entangle the qubits you need first & use quantum teleportation? I mean, I'm no physicist...
  • by Hognoxious ( 631665 ) on Saturday January 21, 2012 @06:34PM (#38777289) Homepage Journal

    I'd run TFA through babelfish if I could work out what language it's supposed to be.

    • by gmhowell ( 26755 )

      For some reason, the writing sometimes seems odd to those who aren't living four simultaneous days.

  • Blah blah (Score:4, Informative)

    by qqe0312 ( 1350695 ) on Saturday January 21, 2012 @06:36PM (#38777297)
    I work in this line physics, which is really a lot of fun. But: The idea that a practical quantum computer will be around in the foreseeable future is jus plain silly. To worry how such a device would be integrated with the cloud is just absolutely bonkers. What a waste of time this all is.
    • by ScrewMaster ( 602015 ) on Saturday January 21, 2012 @08:16PM (#38777767)

      What a waste of time this all is.

      Yoda ... is that you?

    • by jlugert ( 698711 )
      This paper comes out of one of the best, possibly the best, experimental labs in quantum optics. A group of people who have knocked over one after another obstacles in this field. Some of which were considered to be not possible in the foreseeable future if I recall. I don't work in this line of physics, I'm not a mathematician, but I have bothered to read the paper. It's not at all bonkers. A useful property of quantum computers was uncovered. As mentioned in the paper, it was only recently that anyt
      • A practical quantum computer already exists, as demonstrated in this result. A tiny computer, to be sure, but perfectly adequate for this demonstration. I doubt anyone knows with much certainty in one direction or the other how long it will take to scale up the bit width of these boxen--this being the current major obstacle.

        Not really. Quantum computers are only practical in that scientists and engineers have managed, through herculean efforts, have made simple machines that can solve utterly trivial computing problems for which no apparatus whatsoever is necessary or desired.

        A practical quantum computer would be a quantum computer suitable for solving practical computing problems on a basis that's competitive with other types of computers.

        Maybe engineers will figure out how to make such computers in the next few decades, but

        • by jlugert ( 698711 )

          Not really. Quantum computers are only practical in that scientists and engineers have managed, through herculean efforts, have made simple machines that can solve utterly trivial computing problems for which no apparatus whatsoever is necessary or desired.

          A practical quantum computer would be a quantum computer suitable for solving practical computing problems on a basis that's competitive with other types of computers.

          Correct--quantum computers are not the next FTP server. Nor are graphics cards much good for Excel.

          Maybe engineers will figure out how to make such computers in the next few decades, but maybe not. Meanwhile, the capacity of conventional computers continues to push out ahead.

          Such authoritative predictions are littered throughout the commentary on the field. Perhaps produced by those uncomfortable with a technology built upon a mysterious process for which there is no authority. More useful than additional replicas, would be to provide a solid basis for these exuberant predictions. Somehow, despite the best efforts of the critics, the tech is picking up steam in overcoming showsto

    • Wait a second -- does that mean that by the time this technology matures, the cloud would have precipitated into an ancient memory?
  • by Anonymous Coward on Saturday January 21, 2012 @06:44PM (#38777341)

    Must resist...want to invest...

    • by Ouchie ( 1386333 )

      Investment at this point makes sense if you have money to throw away like DARPA. There are few actual functioning quantum computers and most are experimental concept models only. I don't see the applications of quantum computing getting past the education, institution, research and governmental use any time soon. If you were able to get a functional array of quantum computers you would likely not find a shortage of paying customers at this point, and your price point would be considerably higher than your

  • by wbr1 ( 2538558 ) on Saturday January 21, 2012 @06:47PM (#38777359)
    ...the same time we get the following:
    • Cold Fusion
    • Flying Cars
    • World Peace
    • Intelligent Patent and Copyright laws
    • An end to hunger
    • A manned mission to Mars
    • Hell, a manned mission to anywhere beyond LEO
    • When samsaric existence ends and everyone attains Buddahood
    • Re: (Score:3, Funny)

      by tiffany352 ( 2485630 )
      Don't forget Half-Life 3.
    • And the rise of the Linux quantum desktop and five years later the Hurd.

      • Not all is always unbreakable, by example, a quantum virus could break the quantum operating system (ej, enjoyingly the Quantum Windows 666) due to the presence of its quantum bug.
        This quantum bug is the quantum decoherence that many quantum researchers suffer from this quantum anomaly.

        The quantum patches (aka, the Quantum Service Packs) don't solve all this problem. The bug decoherence may always exist, it's a defined property of the Quantum Theory, and could be minimized its presence under many certain qu

    • Hell, a manned mission to anywhere beyond LEO

      Didn't people get that a few times around the 60's?

    • Comment removed based on user account deletion
    • ...the same time we get the following:

      Cold Fusion

      Check. [wikipedia.org]

      Flying Cars

      Street cars kill enough people as it is. Flying cars would be like World War II every year.

      World Peace

      We already have world peace. It's just that some countries don't want to participate.

  • by Chemisor ( 97276 ) on Saturday January 21, 2012 @07:04PM (#38777419)

    Those who do not understand mainframes are bound to reinvent them. Poorly.

  • I really think we should come up with a team to figure out how in-home cold fusion reactors can be integrated into the existing power grid. This is a pressing issue, and I know if we work together we can achieve seamless integration. How's 6pm on Thursday sound?
    • by Anonymous Coward

      I really think we should come up with a team to figure out how in-home cold fusion reactors can be integrated into the existing power grid. This is a pressing issue, and I know if we work together we can achieve seamless integration.

      How's 6pm on Thursday sound?

      Sounds great! I'll bring the salsa & chips ( my salsa's a bit spice-wonky though: You have to stir it really well, or you get hot spots).

  • by hackus ( 159037 ) on Saturday January 21, 2012 @07:38PM (#38777595) Homepage

    I like this quote: "It's good to know that quantum computers will be secure when they exist."

    Gotta love slashdot, ya know.

    First of all, if we ever get a real working quantum computer...and that is a gigantic _IF_ in caps, you can rest assured that someone will break it.

    I would be very surprised if it couldn't be cracked.

    -Hack

    • There are real working quantum computers out there. They are just not powerfull enough to be usefull.

      Also, "secure" on the phrase you quote has a completely different meaning from what you use on your comment.

  • by gweihir ( 88907 ) on Saturday January 21, 2012 @08:54PM (#38777975)

    Currently there is no reliable indication that a) quantum computers of sizes that perform better than traditional computers are feasible engineering-wise and b) that the physics holds up.

    While Quantum computing certainly has captured hearts and minds, at this time it is merely a dream, and one that quite likely will not come true. Incidentally, many experts in the field admit this, but not publicly as that would jeopardize their funding.

  • I'm writing this from my iTab with Retinal Display which has no local storage aside from protein based RAM, since bandwidth is instantaneous and cheap with Data Teleportation so everyone stores their data on Facebook's Quantum Cloud Units, known as FQCU. In fact quantum computing combined with data teleportation is so fast that we found that time travel is indeed possible, for our data communications at least, which is what I'm doing right now.

  • I think the word "perfect" is too strong. A system is only as good as its weakest link.

    Quantum systems are not able to provide any guarantees WRT to *what* the system is entangled with.

    You still need "classical" source of trust to bind the quantum system to do anything useful. (See MITM..quantum proxy server..)

    Questions:

    What prevents the replacement of the quantum cloud service with an attacker who intentionally provides wrong answers?

    Or simply ignores a request pretending they did not get it in a bid to

  • by internet-redstar ( 552612 ) on Sunday January 22, 2012 @06:02AM (#38779825) Homepage
    Quantum computers... will probably work.
    But I guess I'll stick with certainty...
  • For the same effect on classic hardware, you might want to look at https://sharemind.cyber.ee/ [cyber.ee].

    Yes, I do have some friends in the project team.

UNIX is hot. It's more than hot. It's steaming. It's quicksilver lightning with a laserbeam kicker. -- Michael Jay Tucker

Working...