Chief Technology Officer, Xanadu HardwareZachary Vernon pointsHe emphasized that silicon photonics is the fastest path to realize fault-tolerant quantum computers that can operate at room temperature, which has a far-reaching impact on the development of quantum computers and provides important support and possibilities for building more stable, reliable and efficient quantum computingSilicon photonics, the technology for manufacturing integrated photonics on CMOS platforms, has become a buzzword for the past two years because the technology promises to provide data centers with faster, safer, and more efficient solutions to cope with the pressure of the increasing transmission demand for artificial intelligence. However, the potential of silicon photonics is not limited to traditional computing and communications. xanadu is a quantum computing company founded in 2016 and headquartered in Toronto, Canada. The company is building fault-tolerant computers based on silicon photonics chips. Using photons as qubits, Xanadu believes silicon photonics will be the fastest path to a fault-tolerant quantum computer that can operate at room temperature. Zachary Vernon, CTO of Xanadu Hardware, spoke to DigiTimes Asia about the opportunities that photonics brings to quantum computing when he attended the 2023 Asia-Pacific High-Level Forum hosted by the Global Semiconductor Alliance last November. 
xanadu cto zachary vernon
The race to achieve fault tolerance
At present, several types of quantum computers based on different principles have been developed, including superconducting qubits, quantum dots, ion traps, and photons. "The difference between the different types of quantum computers is the use of different types of hardware," Vernon noted. Currently, there is fierce competition between them, and different types of quantum computers are suitable for solving different problems, but all are in the prototype stage and cannot be commercially applied. Ideally, if all the methods for building a quantum computer are successful, then they should be equivalent and able to solve the same problem. "To build a more reliable and stable quantum computer, we need to be fault-tolerant and error-correcting. When it comes to achieving fault tolerance, both scalability and performance are important. This means that in a quantum computer, a large number of qubits are needed to encode and correct errors, and these qubits need to be of high performance. Currently, photonics is considered to be the fastest way to achieve this and scale. ”
Photonics makes it possible to network different chips in different ways using optical fibers, which provides better connectivity than superconducting methods. "Because photonics provides better connectivity, it means that some quantum coding techniques, particularly quantum low-density parity (LDPC) error correction coding, can be better designed and implemented to improve reliability and accuracy," Vernon said. In fact, photonics is the only way to achieve this capability. Because other methods are limited in connectivity between qubits, photonics can use optical fibers to route and transmit qubits within a quantum computer or network to a desired location or node, so that quantum information can be transmitted and processed more efficiently. This capability is critical for building large-scale quantum computers, as it allows qubits to communicate and exchange in a more efficient and flexible way, better supporting the computation and functionality of quantum computers. As the number of qubits increases, so does the importance of photonics in quantum computing. To take advantage of photonics, quantum computers may require millions of qubits. Xanadu's ability to leverage better LDPC technology and access ten to a hundred times more logical qubits than competing methods gives it a greater potential advantage in quantum computing. Vernon emphasized: "We think anyone working in silicon photonics manufacturing should keep a close eye on the quantum computing industry. ”
Photonics will be the fastest way to scale
According to Xanadu**, once fault tolerance is achieved and scale-up begins, adding hundreds of logical qubits per year, Xanadu alone will require hundreds of thousands of 300 mm wafers per year. Because a quantum computer is like a data center that actually requires thousands or millions of chips to build. This is a significant market opportunity where the yield of silicon photonics wafers required for quantum computers will be compared to the production of silicon photonics wafers currently being produced in a few years. "In the coming years, Xanadu hopes to be fault-tolerant and scale to 1,000 error-correcting logic qubits," Vernon said. This looks like a data center with about 10,000 racks. Currently, the company's focus is on developing the hardware needed to deliver fault-tolerant computers deployed in the cloud. "In the long term, the use of silicon photonics technology has the potential to deploy quantum computers close to the edge terminals. "In principle, quantum computers using photonics can be placed inside consumer devices. "Zachary Vernon explains"Although certain technologies need to be developed, in principle, such capabilities exist, since they can all work at room temperature. But whether it is actually feasible will take more time to study. "From a market perspective, Pennylane, Xanadu's software library for programming quantum computers, is the company's main product. The product was developed in collaboration with NVIDIA and Amazon Web Services. "Pennylane is one of the leading software APIs for developing algorithms for quantum computers," commented Zachary Vernon, "and it started out specializing in machine learning applications: quantum machine learning, and has developed a community around it that accounts for a sizable portion of the algorithm development market." Xanadu's CTO also highlighted Pennylane's hardware-agnostic nature: it's not limited to photonic quantum computers or Xanadu's hardware, and it can be used on different platforms. At present, Xanadu has cooperated with several hardware vendors to confirm this. For example, Xanadu has partnered with several automakers, including Volkswagen, to develop quantum algorithms for battery simulation using Pennylane. 
The risk of missing out on global competition
With the AI revolution on the horizon, Xanadu noted that "although the field of quantum machine learning is still in its infancy, a lot of research work has already been done. In the vast array of algorithm developments currently underway, quantum computers seem to be able to handle certain machine learning tasks in very different ways. However, there is still a need for large-scale, fault-tolerant quantum computers before their principles can be fully explored. Once these technologies scale up to very large scales, they can more efficiently handle traditional machine learning basic operations, such as matrix operations. Vernon argues that quantum computers will not replace data centers, but will augment them, suggesting that quantum computing does not solve the computing problems and applications that are currently being done by edge clusters. "Due to the mathematical nature or structure of the problem, there are some situations that ordinary classical computers cannot handle, and the unique approach adopted by quantum computers is able to effectively solve such problems," he noted. Moreover, the development of quantum computers is not a simple improvement or minor optimization of existing traditional technologies. A good example of this is Borealis, the latest cloud-deployed machine, which is capable of beating the world's most powerful supercomputers and is orders of magnitude more efficient than Fugaku. "Fundamentally, quantum computing deals with a completely different set of problems that can't be solved by simply expanding data centers. "It opens up a market for applications that are inaccessible with existing technology, and it will always be. ”
Given the strategic importance of quantum computing, a global competition has begun. Commenting on Canada's strengths, Vernon said that the country has excelled in the ecosystem, especially when it comes to talent development, with many of Canada's leading physicists and engineers being hired directly by Xanadu. When talking about Taiwan's advantages, Vernon believes that Taiwan is the "Mecca of semiconductors" and therefore will also become the center of silicon photonics in the future and play a key role in the future ** chain of Xanadu. With the help of the Canadian office in Taipei**, Xanadu has established relationships with a number of wafer fabs and assembly and test facilities in Taiwan. For other photonics ecosystems, Vernon sees silicon nitride and lithium niobate as two extremely important emerging platforms that Xanadu has been working on for quite some time. "If you don't get involved, you're missing out," Vernon said. Over the past few years, the United States and Europe have spent a lot of time and effort on photonic quantum computing, so they are already slightly ahead. He hopes that all countries will clearly understand the potential power and importance of quantum computing for the future, and actively devote themselves to the field of quantum computing. (This article is based on DigiTimes Asia and is only used as industry information and news sharing).
Luceda China Profile: Luceda Photonics is a provider of optoelectronic integrated design automation software spun off from the IMEC Microelectronics Research Center in Belgium, Ghent University, and the Université Libre de Bruxelles. As a leading enterprise in the field of photonic integrated design automation software, we are committed to providing customers with full-process software and services for optoelectronic integrated chip design, PDK construction and operation and maintenance, and assisting optoelectronic integrated design engineers to enjoy the same "first-time-successful" design experience as electronic integrated design engineers.
In 2021, luceda China landed in Shanghai, providing Chinese customers with more efficient and closer to customer services, adhering to the people-oriented concept and open and win-win attitude, using cutting-edge professional technology and good design experience to help improve product quality, shorten product launch cycle, cultivate core competitiveness, and promote the rapid development of China's optoelectronics and related fields.