Quantum algorithm DNA computing and post-classical computing era 【Full Text】

Security exhibition network technology dynamic binary combined with the great computer, promote humanity into the information age. In this new world based on the material world, consisting of 0 and 1, we rely on algorithms and electronic technology to continuously solve a large number of problems that could not be solved.

However, curious humans are always good at asking new and more complex questions, which in turn drives the advancement of computing technology. This time, we started to be at a new node, that is, with the gradual failure of Moore's Law, we will usher in the era of post-classical computing. In this era, quantum computing, biological computing, etc. will begin to take the stage of history.

These two new algorithms will help us solve problems that now seem difficult to solve. Although they are both in the early stages of development, their exploration and continuous progress deserve our long-term attention.

1. Where to go for classic computers

There is no doubt that the Internet is the embodiment of classic computing power. Billions of computers of various shapes and sizes around the world form a network through algorithms, radio signals, and fiber optic cables, and collaborate with each other to create a special way of life in the universe as we know it. What is even more incredible is that classical computing completed this feat in less than two generations. This is a rate of technological progress without historical precedent.

Behind this progress, Moore's Law proposed in 1965 has always played a theoretical role. But under this law, silicon computer chip is a physical material after all, so it is subject to the laws of physics, chemistry and engineering. When we shrink the transistors on integrated circuits to the nanometer level, the transistors can no longer be reduced every two years as usual.

Take Intel ’s continuous exposure of security vulnerabilities in its processors as an example. To a certain extent, this is because engineers must try their best to improve the performance and speed of the processor, which is physically impossible to improve the integrated circuit itself. .

As transistors shrink to only 7 nanometers long, engineers have reached the point where transistors use a small number of atoms to make working elements. The structural integrity of any smaller transistor will quickly collapse and lose the ability to control and direct current, and it is the information passed by the current that makes the computer so powerful.

When the conversion and control of current is improved, the computer can be faster and more flexible. However, you cannot let an electron move at a speed that exceeds the speed determined by the medium through which it passes. The way to "accelerate" the flow of electrons is to reduce the distance it moves between logic gates, and the result of this operation can be several trillions of a second faster than before, which is what we have been doing for 40 years. thing.

The processor of a classic computer is undoubtedly fast, but unfortunately it is not fast enough. Despite the incredible capabilities of the classic computer, it is still in the face of difficult but extremely important mathematical problems (such as optimization and protein folding). The sequential nature of classical computer operations means that it can never catch up with the growth rate of an O (2n) or O (n!) Problem.

No one is willing to accept that the incredible technological journey we have enjoyed over the past half century is coming to an end, but unless we find an algorithm that can provide a shortcut to this growth rate, we must go beyond classical computers.

2. The advent of quantum algorithms

In the 25 years since Peter Shor published a quantum algorithm (quantum algorithm for decomposing large prime factors), mathematicians and computer scientists have developed other quantum algorithms to solve problems that are difficult to solve in classical computers.

Among these dozens of quantum algorithms, many are orders of magnitude faster than the effective classical algorithms we know. Of course, these algorithms can only be implemented in their unique quantum environment.

Some important work in the field of quantum computing is to create algorithms that simulate various quantum systems, ranging from laser technology to medicine. These algorithms will largely surpass similar classical computational simulations. At present, the classical algorithm for molecular simulation is based on the types of molecules it can simulate. These algorithms are usually limited to molecules with less than 70 spin orbits, and the complexity of the simulation grows so fast that it becomes more and more difficult to handle.

And a qubit can effectively represent one of these orbits. A quantum computer with only 100 qubits will be able to perform molecular simulations that are unmatched by classic computers. These simulations may reveal various previously unknown compounds and may provide new treatments for various diseases.

From depth-first search to adiabatic optimisation, quantum algorithms are widely used and are constantly improving. When these algorithms are really put into use, some frustrating, difficult, and exponential problems in business, administration, medicine, engineering and other fields will be solved. However, what these algorithms lack is a quantum computer with enough qubits and powerful enough.

Overall, quantum computing technology is still in its infancy, which not only involves the qubits you must master, you must also discover a material that can superconduct at room temperature, and figure out how you can maintain the internal environment of the qubits so that Work as close to zero as possible.

In addition, the vast majority of work that a computer needs to do will not be faster on a quantum computer than on a classical computer, because sequential operations are not designed for quantum computers. We will still use classic computers for a long time after quantum computers are fully available, and quantum computers may be placed in enterprises and national laboratories to provide processing services through cloud computing.

3. Redefine computers for the post-classical era

The problems faced by classic computers are inherent in the electronic nature of the computer itself. The computer evolved from a simple electronic circuit and used a very specific calculation method to solve the problem, so it was locked in the continuous binary number calculation model that electronic technology has used for more than a century. But the dominance of this model in our current technology does not mean that it is a method of performing calculations.

We can look away from our obsession with silicon chips and take a look at another major area of ​​computing research: DNA computing. This is an area with incredible development potential. This concept may seem a little strange at first glance and messy. But if you think about it carefully, it is obviously a candidate technology for post-classical computing research and development.

DNA coding has become a powerful data transmission and storage mechanism, but researchers are now digging deeply into the various components of DNA itself, and it may become a computer system itself.

Studies have shown that four different amino acids (A, T, C, and G) serve as the building blocks of DNA and can be reused as codeable bits. When mixed, these amino acids naturally self-assemble into DNA strands, not just any DNA, but the possible DNA arrangement of all available materials.

This is a game-changing innovation, because performing operations on the superposition of qubits is different from true parallel computing. Quantum computers will only give you a single output, either a value or a resulting quantum state, so their effectiveness in solving exponential or factorial time complexity problems depends entirely on the algorithm used.

However, DNA computing takes advantage of these amino acids' ability to construct and assemble long-chain DNA. Mixing these amino acids, they will naturally form a longer and more complex arrangement of amino acids. The arrangement is all about optimization, and even a quantum computer is likely to find that this optimization exceeds its capabilities.

This is why DNA computing is so exciting. The ongoing research on DNA computing will reveal its true efficacy in time, but the self-assembled DNA chain provides the prospect of true parallel computing, even quantum computing cannot claim this.

Overall, whether it is quantum computing or DNA computing, they will redefine what we know about computing, and we will integrate these different models to create new systems and have a lasting impact.

Although it may be interesting to speculate on specific progress, what is more important than any progress is the synergistic effect of these different progresses, such as blockchain, 5G network, quantum computer and artificial intelligence.

(Reference information: interestedengineering)