Hybrid Quantum-Classical System from D-Wave and Where Quantum Computing Could Be Used

D-Wave Systems announced the availability of D-Wave Hybrid™, a simple open-source hybrid workflow platform for building and running quantum-classical hybrid applications. D-Wave Hybrid is part of the Ocean software development kit within D-Wave’s Leap™ quantum cloud service and is available today for download.

Quantum and classical computing will work together. Classical computers are easy and best for simple work and quantum is only needed where it is superior to classical.

We waste classical computing resources for ease of use. Most people do not use GPUs to solve computationally intensive problems 10X to 100X faster.

Quantum computers will be for ultra-hard problems where they are needed.
They could also be used where true random numbers are needed and if there are some killer quantum algorithms created to make all large database searches faster. They could become the accelerator for overall search and other ultra-large applications.

There are also optimization problems for routing planes and national and global systems. Fedex and other large organizations could have logistics made more efficient.

There could be advantages for machine learning and drug discovery.

We could all benefit indirectly from the advancement of science and more efficient logistics.

D-Wave Hybrid includes:

* Hybrid workflow control: enables rapid development of hybrid applications that can run across classical, the current D-Wave 2000Q™ family, and future quantum systems
* Modular approach: incorporates logic to simplify distribution of classical and quantum tasks, allowing developers to interrupt and synchronize across the systems and draw maximum computing power out of each system
* Problem deconstruction: capable of breaking down large problems that are bigger than the quantum processing unit (QPU) into parts that are then recombined for the overall solution
* Familiar coding environment: familiar Python-based framework and documentation includes examples that make it easy to get started without knowledge of quantum mechanics
* Flexibility: includes a number of example hybrid workflows, allowing developers to explore which workflows are best for the problem they are solving
Leap quantum system access and Quantum Application Environment (QAE): free, real-time access to the D-Wave 2000Q family of systems and QAE resources, including learning tools, and community and technical forums for easy developer collaboration

2 thoughts on “Hybrid Quantum-Classical System from D-Wave and Where Quantum Computing Could Be Used”

  1. …1000x time greater hardware cost….

    Dwave realized a long time ago, they will not be able to keep the lights on in the long run by just selling units of their Qcomputer. You can save a lot of time by developing your software on classical hardware and execute it in their QCloud, you just pay for that time slice. You can rent cloud time on a classic supercomputer for under $2k an hour, but the cost will outstrip QCloud if your classic job runs for many hrs or a few days.

    Run time savings has value.

    Classical computers just having optimal solutions to the same problem isn’t the only consideration.

  2. A couple of thoughts:

    1. Nobody needs quantum computers to solve a problem when classical computers have solutions to the same problem that are on the border of optimal. The quantum machine will provide basically the same solution with a 1000x time greater hardware cost. Yes there are problem domains where you get quite close to optimal (and sometimes the optimization part isn’t the bottleneck).
    2. Quantum computers will do well for future optimization problems and for areas where we don’t have good heuristic solutions (e.g. protein folding).
    3. We were promised quantum computers and we got hybrid-quantum computers. Basically the quantum computer generates a good starting set for the classical computer to finish optimizing. This is nice but it doesn’t get around needing classical computers and their custom optimization software.
    4. One does not simply deconstruct larger problems in to smaller problems. If you could do easily or effectively this then it begs the question of why you need more qbits.

Comments are closed.