Please note that JavaScript and style sheet are used in this website,
Due to unadaptability of the style sheet with the browser used in your computer, pages may not look as original.
Even in such a case, however, the contents can be used safely.
- NEC's R&D
- R & D Domains
- Global R & D
- Press Release
- Innovative Engine
- Speech Recognition
- Quality degradation analysis for large-scale networks
- Bioplastics for Electronic Equipment
- Quantum Computer
- What quantum computers can do for us
- What does "quantum" mean?
- Basic principles of quantum computers and differences with existing computers
- Beginnings of quantum computer research
- Qubit solid-state device: 1-bit operation
- Quantum computer operation mechanism
- Qubit solid-state device: Demonstration of a C-NOT circuit
- Demonstration of Coupling-controlled Qubits and Future Outlook
- About Dr. Tsai & Mr. Nakamura
- Carbon Nanotube
- Awards
- Access
Quantum Computer
What quantum computers can do for us
I am Jaw-Shen Tsai of NEC's Green Innovation Research Laboratories, and I would like to tell you about some of the fascinating things that the quantum computer will make possible. I will also explain what "quantum" means.
To begin with, there are still many issues that must be solved before a practical, working quantum computer can be achieved. A quantum computer is not something that we can expect to be using any day soon. The completion of a quantum computer, however, will make it possible to instantly solve problems that require exhaustive calculations by "brute force," a method that existing computers are not particularly strong at. Problems of this type include the computing of prime factors and the solving of NP-complete problems that must consider the combination of a vast number of events. We can also expect the quantum computer to be applied to quantum simulations for analyzing complex quantum systems such as protein reactions and catalytic material, and to secure communications based on quantum cryptography.
At the same time, it is sometimes mistakenly thought that any type of computation will be accelerated by a quantum computer. This misunderstanding originates with the false idea that a quantum computer achieves high speed in the same way that a personal computer would by increasing the speed of its CPU. Let's take the case of calculating 1+1 in a quantum computer. The time required for performing this simple calculation and outputting its result would not be very different from that of an existing computer. In short, the wonderful features of the quantum computer will probably not be appreciated as long as we think of it in terms of the computers that we now use in our daily lives. However, for specialists that are working on problems adequate for the quantum computer and that wish to perform computations that would require a huge amount of time even for a supercomputer, the quantum computer is a dream come true.
The following graph shows how existing computers and quantum computers are expected to evolve in terms of integration and information-processing capacity. According to Moore' Law, which states that data density in semiconductor devices doubles every 18 months (an empirical law first observed by Gordon Moore in 1965), the length of high-speed semiconductor gate electrodes in existing computers should reach 6 nm by the year 2020. Concerning this area of research, NEC announced the successful development of a 5-nm transistor in December 2003. A transistor size of 5 nm, however, is thought to be the limit of device miniaturization in terms of operating principles. The quantum computer, on the other hand, will eventually surpass the information-processing capacity of existing computers as the number of quantum bits steadily increases.
What is an NP-complete problem?
This is a problem for which an optimal combination cannot be immediately determined.
A typical example of an NP-complete problem is the "traveling salesman problem" in which the shortest route for visiting multiple destinations must be found. The problem of determining what sales route a salesman should take to make his sales activities most efficient may appear to be simple at first. But if several tens of destinations must be visited, even a supercomputer would need several hundred million years or more to solve this problem. A method for calculating a solution to this problem in an efficient manner has not yet been found.
What is the current capacity of supercomputers?
NEC's SX-9 series of supercomputers has a peak processing capacity of 100 GFLOPS per processor. --> SX Vector Solutions(http://www.necam.com/sx/)
"FLOPS (floating-point operations per second" is a unit of computer processing speed. A processing speed of 1 FLOPS means that the computer in question can perform one floating-point operation per second. Accordingly, 1 GFLOPS means a billion floating-point operations per second. A "floating-point operation" may also be called "real-number arithmetic" as opposed to "integer arithmetic."
