Slashdot Mirror
D-Wave Previews Quantum Computing Platform With Over 5,000 Qubits (venturebeat.com)
An anonymous reader writes: D-Wave Systems, one of the handful firms that is building a quantum computer, today unveiled the roadmap for its 5,000-qubit quantum computer. Components of D-Wave's next-generation quantum computing platform will come to market between now and mid-2020 via ongoing quantum processing unit (QPU) and cloud-delivered software updates. The complete system will be available through cloud access and for on-premise installation in mid-2020.
They generate hype and sweet sweet VC dollars?
Unless things have improved, their systems are still slower and more expensive than solving the same problem on general digital computers, and probably still slower than using analog computers.
... the stars will all start going out.
CUR ALLOC 20195.....5804M
As far as I know, D-Wave doesn't make any universal quantum computers, but only quantum annealers. That means that they can solve some optimization problems on their machine, but they can't actually run e.g. Shor's integer factorization algorithm. As far as I'm aware, the current record for universal quantum computers is Google's Bristlecone, which has 72 qubits with a single-qubit error rate of ~0.1%. For comparison, most quantum error correction require an error rate of below 0.001% or so, and running Shor's algorithm to break 2048-bit RSA encryption might require up to 10,000 qubits. It'll probably be a while until they'll find those primes for you.
There are two types of machines called quantum computers. The ones with 6-15 qbits are using special "entangled" logic gates that can in theory solve any problem, but getting a usable circuit is very difficult. D-wave is a quantum annealer, meaning that it is meant to solve a particular kind of optimization problem. It's kind of like putting a bunch of odd shaped items into a box and shaking the box. The random movement will over time cause the items to pack more tightly into the box. Only in this case, the "shaking" is quantum fluctuations.
the article wasn't very clear on this.
Yes, they are quantum annealers that are capable of doing discreet optimization problems.
Protein folding, travelling salesman, quantum chemistry modeling, and artificial neural networks can all benefit from them.
They will not run Shor's algorithm and crack encryption but the class of problems they can tackle is still has very real world applications.
The number of qubits is increasing from 2000 to 5000. The connectivity is also increasing from 6 couplings per qubit to 15 couplings per qubit. The connectivity is important because it allows more complex problems to be solved. For example, more qubits means you can solve problems that have more variables. But you can't solve any interesting problems unless the variables interact with each other. More connectivity means you can solve problems with more interactions between the variables.
D-Wave processors are quantum annealers, which is a type of quantum computer. You can think of it as having a reduced instruction set relative to gate-model quantum computers. Quantum annealers _can_ be universal, but D-Wave's aren't yet. They also have noise issues but are far more robust to noise than gate-model quantum computers. Right now D-Wave processors provide a heuristic algorithm (quantum annealing) for an NP-hard optimization (Ising minimization, equivalently, quadratic unconstrained binary optimization (QUBO)). Quantum annealing does, in theory, scale better than classical alternatives like simulated annealing. There has been some evidence of this (https://arxiv.org/abs/1705.07452) and I'm sure more will be coming.
What are the problems that D-Wave has so far managed to solve, that a conventional computer cannot solve just as efficiently and at a fraction of the cost?
They will not run Shor's algorithm and crack encryption but the class of problems they can tackle is still has very real world applications.
In fact, the factoring problem has a particularly nice algorithm in the adiabatic regime. It's just a multiplication circuit, where you clamp the outputs and "run it backwards" to deduce the inputs. It's not Shor's algorithm, but D-Wave is way ahead of the (gate-model quantum) competition in terms of factoring.