System architecture in quantum computing – history repeating itself?

A practical quantum computer is closer to reality than ever before. Massive investments of time and resources are accelerating research in quantum computing, bringing us steadily nearer to viable implementations. Optimistic projections suggest that a usable quantum computer (QC) could emerge within five years ​[1]​, while one capable of breaking conventional cryptosystems such as RSA and Diffie-Hellman may be around a decade away ​[2]​. In this article, we examine the current state and challenges of quantum hardware, explore the evolution of system architectures for QCs, and identify areas where intellectual property could play a pivotal role—revealing how history may be poised to repeat itself.

Where quantum hardware poses challenges, system architecture unlocks opportunities

The discussion around quantum hardware has primarily revolved around physical implementations of qubits. This is the fundamental building block of quantum computing, the quantum counterpart to a classical bit. Unlike a classical bit, which can only take one of two discrete values (0 or 1), a qubit exists in a continuous state space, demonstrating phenomena such as superposition and entanglement ​[3]​. In principle, any physical system with two distinct quantum states can serve as a qubit, though some implementations are more practical than others. Today, the most widely adopted approaches include superconducting qubits (notably linked to the 2025 Nobel Prize in Physics as discussed in an earlier article), trapped-ion qubits, and photonic qubits. This remains an active research area with significant investment from major technology companies such as IBM, Intel, and Google. Despite progress, substantial challenges persist—particularly in scaling systems, achieving reliable read/write operations, and performing computations before quantum states degrade due to noise and decoherence ​[3]​.

From a historical perspective, quantum computing today is roughly at the stage classical computing was in the mid-1930s. The first patent for a transistor, a fundamental hardware component in classical computing, claims priority in 1925 and was invented by J.E. Lilienfeld (US Patent 1745175 A). However, it was not until 1948 that Bell Labs produced the first practical semiconductor transistor, enabling commercial applications. Quantum computing currently sits between the initial demonstration of a physical qubit and the onset of large-scale commercialisation—the modern equivalent of 1935. Later, in the 1960s, the concept of an instruction set architecture (ISA) was introduced in classical computing, which provided a translation layer between hardware logic and low-level software. This sparked a debate: should ISAs remain minimal, leaving complexity to software (RISC), or should hardware implement complex instructions for performance (CISC)? Both approaches ultimately succeeded. RISC architectures such as ARM dominate mobile devices, while CISC architectures like Intel’s x86 power desktops ​[4]​. A similar discussion is likely to emerge in quantum computing, and this is an area where intellectual property could once again play a significant role.

A quantum ISA

Unlike classical computing, where ISAs emerged decades after the first hardware breakthroughs, ISA development can already be found in quantum computing. One example is the recently granted US patent 12386604B2, which discloses a quantum ISA by research scientist Anastasiia Butko at Lawrence Berkeley National Laboratory. In the patent, a key distinction is highlighted: classical architectures rely primarily on a single type of device, arithmetic logic unit (ALU), to perform various operations, whereas quantum architectures employ a variety of quantum gates applied directly to qubits ​[5]​. Consequently, the design question shifts from which operations to implement in an ALU to which quantum gates should be natively supported on a quantum chip.

This raises a familiar debate. In quantum computing, a universal set of gates is one to which any quantum computation can be reduced. While it is logical for a quantum processor to have a universal set, whether that set should be minimal remains an open question. The cited patent takes a clear position, proposing that “a quantum device exposes a minimal set of native gates that constitute a universal set” and extending a reduced instruction set (RISC) to a quantum ISA ​[5]​.

The landscape is ripe for innovation

A brief survey reveals only two other patent families directly addressing quantum ISAs, one of which also adopts a RISC-based approach. With just three families identified, this appears to be an area ripe for innovation. Could a CISC-based quantum ISA offer advantages over RISC-based designs, or might both coexist as in classical computing? The architecture that achieves large-scale adoption will likely create licensing opportunities, similar to the dominance of x86 in classical systems.

In summary, quantum computing development is progressing on multiple fronts. While hardware research continues to grapple with the challenges of fundamental physics, system architecture is already gaining traction. The field remains open for innovation, and as scalable quantum computers approach, we may see a quantum version of the RISC versus CISC debate. When that time comes, EIP stands ready with its Quantiphy offering to provide strategic guidance and expertise in intellectual property for quantum technologies.

​​References

​[1]  

​D. Milmo, “Google hails breakthrough as quantum computer surpasses ability of supercomputers,” The Guardian, 22 October 2025. [Online]. Available: https://www.theguardian.com/technology/2025/oct/22/google-hails-breakthrough-as-quantum-computer-surpasses-ability-of-supercomputers. [Accessed 4 December 2025].

​[2]  

​P. Asplund, “Forskar om att framtidssäkra internet,” KTH, 30 October 2024. [Online]. Available: https://www.kth.se/om/nyheter/centrala-nyheter/forskar-om-att-framtidssakra-internet-1.1366813. [Accessed 4 December 2025].

​[3]  

​Wikipedia contributors, “Quantum computing,” Wikipedia, The Free Encyclopedia, 2025.

​[4]  

​S. Sinofsky, “RISC v CISC: An Age Old Debate,” Medium, 3 May 2022. [Online]. Available: https://medium.learningbyshipping.com/risc-v-cisc-an-age-old-debate-79d859668d35. [Accessed 5 December 2025].

​[5]  

​A. Butko, “QUANTUM INSTRUCTION SET ARCHITECTURE”. US Patent 12386604 B2, 12 August 2025.

‍

‍