“2020 was the first year of quantum computer use”
Currently, quantum computers are popular all over the world including, and venture companies all over the world are fighting on the same ground. It’s worth noting that IBM, MS, and AWS have begun to open their doors to the general public. In this article, Yuichiro Minato, CEO of MDR, who has worked on numerous projects, will explain the present and challenges of quantum computers, their future, and how they can be used in financial institutions.
* This article is based on the lecture content of “REBOOT-Money, Payment systems & Financial service” in the event “4F (Future Frontier Fes by FINOLAB)” 2020-REBOOT held by FINOLAB in March. It is a reconstructed one.
Why quantum computers are booming
Mr. Minato runs a venture company (MDR) that develops quantum computers and software, and the company won the second prize at the “3rd MUFG Digital Accelerator” held by Mitsubishi UFJ Financial Group (MUFG). There is.
The business model of the quantum computer industry is very simple, and basically, we are developing applications for each company using software for making quantum computers such as libraries, open-source, and SDK.
Currently, quantum computers are popular all over the world including, and venture companies all over the world are fighting on the same ground. The number of downloads and the number of participating developers is very large. Information exchange among developers is also active, and the situation is exciting.
In addition, the field of quantum computers has a very strong academic color, and I sometimes write treatises while working. On the other hand, regarding how it can be applied to business, the university and the corporate side are working together on a daily basis.
In 2020, various quantum computers appeared
What is a quantum computer? Mr. Minato describes a quantum computer as “a new computer that arbitrarily switches between bits and waves as particles to perform calculations.”
Conventional computers calculate 0s and 1s one by one, but quantum computers superimpose 0s and 1s to convert them into three-dimensional “waves”, and then superimpose the waves in one place for complicated calculations. Are calculated together. It is a mechanism that the calculation is completed if the wave is restored.
Quantum computers can solve large problems
Currently, quantum computers can solve large problems (combinatorial optimization problems) at higher speeds by specializing in some calculations and “general-purpose type” called “quantum gate method”, which is relatively compatible with conventional computers. There is a “quantum annealing type”.
In 2019, there was big news that Google achieved “quantum transcendence (quantum computer transcends the limits of the performance of modern computers)”.
This is the result of confronting the world’s largest supercomputer with a quantum computer with a small chip of 1 to 2 cm square, and the calculation that takes 10,000 years with the supercomputer can be completed in 200 seconds with its performance. It showed sex. However, quantum computers need to be operating at low temperatures. So equipment such as “experimental facilities” is required, and calculations are apt to go wrong, making it difficult for ordinary companies to own (photo).
Therefore, usually, for example, when using D-Wave’s “quantum annealing” quantum computer released to the general public in practice, the method of using a machine located in North America or Canada in the United States via the Internet is adopted. That is.
Until now, the “quantum annealing type” represented by D-Wave has been the main choice for companies, but recently, the “ion trap method” in which atoms are floated in the air and shot with a laser, and the “photonics machine” using optical fibers, etc. Etc., and various new methods have been introduced to the market.
IBM, AWS & MS open the door to quantum computers in 2020
In addition, from 2020, major IT companies will make it easier to contract quantum computers at low prices, and there is a move to open the door to users widely.
For example, IBM is a machine that can use 53 qubits, Microsoft (MS) is Honeywell and IonQ’s “ion trap type”, QCI’s superconducting machine (quantum gate machine controlled by superconducting circuit), Amazon Web Service ( AWS) offers three types of quantum computers: D-Wave, IonQ, and Rigetti in Canada.
Suddenly using a real quantum computer is a high hurdle, so there is also a method of developing by putting a simulator tool in your PC first. If all goes well, you can send the problem to a real quantum computer over the Internet.
Quantum computers are still in their infancy!
Quantum computers are still in their infancy and cannot solve very “large-scale problems”. The procedure is to “shred” the problem, run each in a simulator, and if successful, “small” the problem and throw the calculation into the actual quantum computer via the cloud. In other words, even now, in a sense, if you download the software, you can start development using a “quantum computer.”
Currently, the quantum computer applications that are mainly use, can be roughly divided into four.
(1) “Machine learning” that learns and uses data,
(2) “Combinatorial optimization problem” that selects one of the best choices from a large number of options,
(3) When a material maker develops a new material “Quantum chemical calculation” used for material calculation.
(4) “Cryptographic related”.
Quantum computer with many problems
Mr. Minato pointed out that as an issue of the current quantum computer, “there are so many errors that it cannot be said that it is completed” at this stage. Today’s quantum computers perform calculations in analog form without digitization after creating algorithms. The calculation is mistaken if “error correction” is not realize.
An ideal quantum computer calculates digitally and implements error correction, but the current quantum gate method cannot avoid errors. Therefore, what can be calculated will change. “NISQ: Noisy Intermediate-Scale Quantum” is a machine with many errors.
However, this does not mean that NISQ cannot be used at present. It has become mainstream to use “hybrid calculation (quantum-classical hybrid calculation)” to avoid errors.
This is to calculate while correcting the error by performing a short calculation. With a quantum computer and repeating the work of applying correction with a conventional computer many times. This makes it possible to solve some practical problems.
Given the various advances made this year towards the realization of practical large-scale quantum computers. It is easy to predict that these advances will continue into 2021. Which is almost certain. In addition, one-day companies will announce that they have succeeded in using quantum computers. To solve practical problems that traditional supercomputers could not.
Solving problems that conventional supercomputers could not solve with quantum computers is the next important milestone for the realization of commercially available quantum computers. But that milestone itself does not put cryptography such as RSA or ECC at risk. The decryption of cryptographic algorithms requires large-scale quantum computers, which did not exist in 2021.
Gaining the ability to solve practical problems will facilitate further investment in the development of better quantum computers. This is exactly the virtuous cycle of feedback loops that follow Moore’s Law. Just as computers used to grow exponentially year after year.
Cryptographic migration can be time-consuming and often decades. Therefore, if you want to be prepared for a quantum computer of sufficient size, companies and organizations must start preparing for it now. Standards bodies and security experts are working hard to lay the groundwork for such a transition.
The National Institute of Standards and Technology (NIST) recently held a workshop on “Considerations for the Transition to Quantum Computer Cryptographic Algorithms”. And NIST selected several algorithms for standardization in late 2021. intend to do something. This will expedite efforts towards the implementation of these algorithms.
All advances in 2020 will bring quantum computers closer to practical use. And 71% of IT professionals believe that quantum computing will be a huge threat in the near future. If we haven’t started preparing yet. We will start to lag behind in 2021.
Get ready now
If you haven’t started preparing for quantum computing yet. Make it your new year’s aspiration to deepen and prepare for quantum computing. Quantum computing will put to practical use in the next 5 to 10 years. Scientists have said that for years now, but now it’s more true than ever.
Quantum computing can take up to 10 years to become mainstream. But it’s an unbeatable competition. Basically, this is a competition between the organization and the computer, and there is no time to relax.
DigiCert is working with industry experts!
DigiCert is working with industry experts to build a PKI ecosystem. To prepare for the threat of future quantum-resistant computer cryptography. So that organizations can begin preparing for quantum computers.
In addition, we can develop a quantum computer cryptographic (PQC) toolkit to verify hybrid PQC / RSA certificates. That counter algorithmic decryption by quantum computers.
DigiCert’s PQC toolkit includes:
- Hybrid RSA / PQC TLS Certificate
- FIXED Apache web server / ISARA catalyst server
- Custom Firefox browser