Cisco Systems has officially unveiled a prototype of an innovative network switch capable of solving scalability challenges in the quantum computing field. This announcement holds significant importance not merely for hardware improvement, but for implementing technology that protects delicate quantum states during information exchange between different quantum computers from being destroyed. Cisco is evaluated as holding the key to accelerating the commercialization of quantum computing by focusing on the latter of two development directions: infinitely scaling single equipment versus connecting multiple units to create one massive system.
Currently, quantum computers are beginning to be utilized as coprocessors for solving complex calculations that are difficult for classical supercomputers or problems that take too long to solve. Companies are focusing on solving optimization problems, analyzing molecular structures for new materials and drug development, and developing quantum-resistant encryption technologies, while researchers are focusing on fundamental physics research and developing new algorithms. However, to leap to practical-level quantum computing, the challenge of drastically increasing the number of qubits, the basic unit of information, remains. While the current commercial level is in the hundreds to thousands, experts predominantly evaluate that even expanding to tens of thousands in the future, millions of qubits are needed for performance improvements perceptible in actual industrial fields.
To overcome these limitations, the solution Cisco proposes is connecting various quantum systems into a single integrated network. Like a translator connecting countries that speak different languages, this switch operates at room temperature and has the advantage of utilizing optical fibers widely used in existing communications. By applying Cisco's patent-based conversion technology to relay input and output encoding methods used by different quantum technologies, it allows quantum computers using different methods to exchange data smoothly regardless of the manufacturer. This is evaluated as a core element of universal quantum networking independent of specific hardware methods and is designed to support all major optical-based encoding methods such as polarization, time bin, frequency bin, and path encoding.
However, this equipment is merely a research prototype and not an official commercial product yet, with the next 1 to 2 years being a critical period for verifying the technology's practicality. Cisco has completed system verification for polarization-based methods, but support for time bin and frequency bin methods is scheduled to proceed additionally. Such universality is particularly important in the quantum computing market where industry standards are not yet fully established. Facilities built to match only specific methods risk becoming obsolete rapidly over time, so Cisco's switch enables research institutions and companies to create a flexible environment mixing custom equipment and universal equipment. Cisco has long-term plans to collaborate with IBM and other partners to build the entire stack, including hardware, software, and protocols for quantum networking, to prepare the foundation for running quantum applications.