Q-Fly: An Optical Interconnect for Modular Quantum Computers

TL;DR

The paper addresses the challenge of scaling quantum computers by introducing Q-Fly, a Dragonfly-inspired optical interconnect that minimizes hops and optical components while enabling entanglement distribution across modular quantum nodes. It presents a full-stack design, including three topology variants, a group-switch construction analysis, and a prototype demonstrating two-hop entanglement with fidelities around 0.60–0.76. Key contributions include end-to-end loss/infidelity analyses, a scalable group-switch approach, and a performance projection tool showing near-term viability for moderate-radix switches and scalability to larger systems. The work highlights cross-layer requirements and outlines a roadmap toward production-scale, standardized quantum interconnects across technologies like ion traps and neutral atoms, with potential pathways to transduction and integration into fault-tolerant architectures.

Abstract

Much like classical supercomputers, scaling up quantum computers requires an optical interconnect. However, signal attenuation leads to irreversible qubit loss, making quantum interconnect design guidelines and metrics different from conventional computing. Inspired by the classical Dragonfly topology, we propose a multi-group structure where the group switch routes photons emitted by computational end nodes to the group's shared pool of Bell state analyzers (which conduct the entanglement swapping that creates end-to-end entanglement) or across a low-diameter path to another group. We present a full-stack analysis of system performance, a combination of distributed and centralized protocols, and a resource scheduler that plans qubit placement and communications for large-scale, fault-tolerant systems. We implement a prototype three-node switched interconnect to justify hardware-side scalability and to expose low-level architectural challenges. We create two-hop entanglement with fidelities of 0.6-0.76. Our design emphasizes reducing network hops and optical components to simplify system stabilization while flexibly adjusting optical path lengths. Based on evaluated loss and infidelity budgets, we find that moderate-radix switches enable systems meeting expected near-term needs, and large systems are feasible. Our design is expected to be effective for a variety of quantum computing technologies, including ion traps and neutral atoms.

Figures (12)

• Figure 1: A three-group Q-Fly with the experimental demonstration shown in red. Entangled pairs of photons are generated at the EPPS nodes. One photon of each pair is measured by the single-photon detectors (SPDs). The remaining photons are guided to the switching BSA node, composed of an optical switch (OSW) and a Bell state analyzer (BSA). The BSA and other detectors are housed inside a cryostat, represented by the cylinder.
• Figure 2: Three Q-Fly variants with an example connection between two end nodes located in separate groups. (a) Single-path quasi-half duplex (SPHD). (b) Dual-path quasi-half duplex (DPHD). Dashed orange line shows the second fiber connecting the two groups. (c) Dual-path quasi-full duplex (DPFD). The two end nodes have two interfaces, allowing two paths to be used concurrently, and with fewer switches in the optical path.
• Figure 3: Optical system, analog and digital control systems corresponding to the red portions of Fig. \ref{['fig:q-fly-experiment']}. FBS, fiber beam splitter. FPBS, fiber polarizing beam splitter. ODL, optical delay line. Pol. comp., polarization compensator. Pol. sel., polarization selector. SHG, second harmonic generation (frequency doubler). SNSPD, superconducting nanowire single photon detector. SPDC, spontaneous parametric down conversion. SW, switch.
• Figure 4: Real parts of $\rho_{A_1A_2}$ (left) and $\rho_{A_2B_2}$ (right). While fidelity has declined, entanglement sufficient for a variety of purposes is still present.
• Figure 5: Entanglement swapping performed on photons $A_1$ and $B_1$ after passing through the optical switch (OSW), leading to entanglement between photons $A_2$ and $B_2$.