Google Will Open Access To Its Quantum Computer To Commercial Customers
Google Corporation plans to open access to its quantum computing devices to commercial companies in the near future, which will be able to use it to solve their tasks, according to the company's Technical Director, Hartmut Neven, during a discussion on the Roscongress Foundation's online platform.
"We have already tested this system in different divisions of Google. We already have agreements with several partners who plan to use the power of our quantum computing devices to test their own algorithms," Neven said.
There are only a few dozen functional prototypes of quantum computers in the world that can be used to solve fairly complex mathematical or physical problems. For normal operation, such machines need to be cooled to ultra-low temperatures. In addition, they require a lot of highly qualified engineers to operate them, which limits their scope of application.
Three years ago, IBM, one of the leaders of the quantum race, took a big step toward removing these restrictions. Its employees created an interface that allows you to control the operation of a quantum computer and solve problems on it from the Internet Cloud. As early as 2018, physicists from the United States used this quantum "cloud" to calculate the strength of bonds between nucleons in an atom of deuterium, heavy hydrogen.
The birth of quantum clouds
Google has similar plans, Neven noted. According to him, the American Corporation has already built a special computing center for this purpose in Santa Barbara (California, USA), created the infrastructure and software basis for this system, and is now conducting its internal testing.
"We hope that our project will attract more and more users as we develop more and more powerful quantum processors that will find more and more applications," Neven added.
At the same time, the scientist noted that existing quantum computing machines are not yet very effective in coping with the supposed main goal for such devices – solving optimization problems, including road traffic or cash flows. This is due to the fact that when switching to quantum computers, these algorithms accelerated only slightly. In addition, each cycle of such machines takes too long, a few microseconds.
"In the near future, we will be able to use quantum computers to analyze data obtained by nuclear-magnetic resonance spectroscopes used in the study of the structures of proteins and other molecules, as well as to prepare data for training neural networks. In the latter case, our existing systems are superior to classic computers, and in the first case, we are close to solving this problem, including thanks to the participation of people from Russia," Neven summed up.