Oct 27 2015
Quantum computing has been a hot topic since late 1980s. There has been no full-scale quantum computers as of yet, but there are significant accomplishments. For example, the first quantum computer consisted of only 16 qubits, and the latest one (summer 2015) – more than 1000. Canadian company D-Wave Systems has been producing 128-qubit computers based on quantum annealing since 2011. There are some currently at work at Lockheed Martin.
Although D-Wave One models are pretty limited in their computing capabilities, they are already showing some superiority over existing processors. Professor Catherine McGeoch tested D-Wave One Vesuvius CPU against an Intel. First of the tests was 0.5 seconds for Vesuvius against whopping 30 minutes for Intel; the second one required a special translation program for D-Wave One, so the results were roughly the same; and the multiple task test showed Vesuvius solving 28 out of the 33 tasks in 30 minutes while Intel only managed to crack 8.
What's stopping quantum computers from moving to wide distribution? There are mainly two obstacles: 1) the necessity of a very high degree of computing accuracy and 2) risks of destruction of distortion of a quantum system from external shocks, such as temperature, electromagnetic impulses, etc. The greatest progress in avoiding these pitfalls was made by a joint Netherlands – USA research
group which presented a diamond-based quantum computer. It’s only 2 qubits strong, but its uniqueness lies in its hybrid nature: 1 qubit is the spin of a hydrogen nucleus, and the other one is the spin of an electron.
Quantum cells based solely on electrons in a solid body can compute very quickly but also quickly become disentangled. The nucleus is much more stable. Using it as a data storage unit helps avoid decoherence. This allowed the abovementioned test specimen work under room temperatures – which is, obviously, one of crucial points for the future mass production of quantum machines.
Can diamonds truly become the basis for powerful computing systems? Young employees of the Institute of Physics agreed to give their opinion on the matter. They are Yury Lysogorsky and Denis Zvezdov, both Junior Research Associates at the New Materials for Quantum Technologies Lab.
Y. Lysogorsky: “There is still much time before a really efficient universal quantum computer will be built. Yes, Netherlanders and Americans created a computer working in room temperatures but the main obstacles on the way of creating a powerful quantum computer are still ahead. For example, several hundred qubits are needed to make quantum computers able to compete with the existing ones. Such upscaling is rarely made without incidents. Also, the abovementioned project cannot be called the final layout of a quantum machine. The researchers are currently testing different configurations because none of them has yet been able to meet all the necessary criteria. So the jury is still out on what the final scheme will look like”.
D. Zvezdov adds: «The test run on the 2-qubit machine is the Deutsch-Jozsa algorithm. Along with error detection and correction it’s one of the simplest tasks aimed at testing if the system is potentially fit for a quantum computer. First such tests were run on another quantum system based on organic molecule solutions. Later that system was also used for Shor’s algorithm.
Shor's algorithm is probably of more practical relevance than quantum search because the former requires at least 7 qubits while the latter – only 3 and Deutsch-Jozsa algorithm – 2. This was demonstrated last year on another diamond system.
We can assume that building a computer with more qubits is the next step. What it will be based on is still unclear, though, because diamond systems pose some really tricky questions in this regard».
So what about quantum computing research at Kazan University? KFU makes some steps in this direction together with its longtime partner RIKEN. PhD candidate Nyiaz Beysengulov recently came back from his internship at the Japanese institute. His work had to do with electrons on superfluid helium – another potential choice for quantum machines.