Source-linked AI summary
SeQUeNCe: A Customizable Discrete-Event Simulator of Quantum Networks
Xiaoliang Wu, Alexander Kolar, Joaquin Chung, Dong Jin, Tian Zhong, Rajkumar Kettimuthu, Martin Suchara
TL;DR
Quantum-network architectures and control protocols remain immature, creating a need for accurate, flexible simulation of growing full-stack systems. SeQUeNCe provides a customizable modular discrete-event simulator with hardware and protocol models, and its nine-router photonic-network studies show that weighted memory allocation increases completed requests by 69.1% and throughput by 6.8%.
Problem
Quantum networks are expanding toward multi-node prototypes, but scaling users and communication distance requires improved architectures, hardware, control protocols, and reproducible performance testing.
Method
SeQUeNCe combines a modular architecture, discrete-event simulation, quantum-network hardware and protocol models, and configurable workflows for alternative designs.
Results
Weighted memory distribution improved aggregate network throughput by 6.8% and completed requests by 69.1% on average.
Takeaways & Limitations
SeQUeNCe supports comparisons of quantum-network technologies, experiment planning, validation, and new protocol design through an open-source, extensible simulator.
Takeaways & Limitations
The simulator currently focuses on fiber-based networks and sequential discrete-event simulation, with additional protocols and parallelization planned.
Abstract
from arXiv · showhide
Recent advances in quantum information science enabled the development of quantum communication network prototypes and created an opportunity to study full-stack quantum network architectures. This work develops SeQUeNCe, a comprehensive, customizable quantum network simulator. Our simulator consists of five modules: Hardware models, Entanglement Management protocols, Resource Management, Network Management, and Application. This framework is suitable for simulation of quantum network prototypes that capture the breadth of current and future hardware technologies and protocols. We implement a comprehensive suite of network protocols and demonstrate the use of SeQUeNCe by simulating a photonic quantum network with nine routers equipped with quantum memories. The simulation capabilities are illustrated in three use cases. We show the dependence of quantum network throughput on several key hardware parameters and study the impact of classical control message latency. We also investigate quantum memory usage efficiency in routers and demonstrate that redistributing memory according to anticipated load increases network capacity by 69.1% and throughput by 6.8%. We design SeQUeNCe to enable comparisons of alternative quantum network technologies, experiment planning, and validation and to aid with new protocol design. We are releasing SeQUeNCe as an open source tool and aim to generate community interest in extending it.
1 INTRODUCTION
Quantum networks are advancing toward multi-node prototypes, but scaling users and distance still requires better architectures, control protocols, hardware, and reproducible testing. SeQUeNCe addresses this need with a customizable, modular discrete-event simulator and open-source implementation.
- Scaling quantum networks requires advances in hardware, network architectures, control-plane protocols, and reproducible performance testing.
- Simulation can evaluate architectural tradeoffs, optimize quantum hardware, and support robust control-plane development as networks become more complex.
- SeQUeNCe is a customizable discrete-event simulator that models quantum hardware and network protocols through modular, separately testable components.
- The simulator translates quantum network protocols into state machines and supports alternative architectures, hardware models, and protocol interactions.
- The simulator is released as open-source software for testing new algorithms, protocols, and devices.
- SeQUeNCe includes a modular quantum network architecture, detailed protocol and model implementations, and three representative metropolitan-network use cases.
2 BACKGROUND
Quantum networks differ fundamentally from classical networks because quantum information cannot be copied or amplified, entanglement requires classical coordination, and quantum states are time-sensitive. Fiber networks remain practical but require repeaters for long distances, while satellite links trade reach for weather restrictions, intermittency, and low throughput.
- Quantum teleportation transfers an arbitrary quantum state using pre-established high-fidelity entanglement and classical communication.
- Fiber links use existing telecommunications infrastructure but incur approximately 0.2 dB/km loss and require quantum repeaters beyond metropolitan distances.
- Satellite links can connect nodes separated by 1,000 km with one station but face weather restrictions, intermittent operation, and low throughput.
- The no-cloning theorem and transmission loss prevent straightforward amplification of quantum signals over long-distance links.
- Quantum networks use entanglement for teleportation or error correction, with classical heralded signals coordinating operations that users cannot directly observe in qubits.
- Quantum operations require precise timing because Bell measurements need synchronization and stored quantum information decoheres over milliseconds to seconds.
3 SYSTEM DESIGN
SeQUeNCe uses a reconfigurable, modular architecture and a discrete-event kernel to model quantum networks with realistic states, timing, hardware, protocols, resources, and applications. Its modules communicate through configurable workflows for requesting, allocating, generating, and consuming entanglement.
- 3.1 Quantum Network Simulation Requirements: SeQUeNCe models quantum states and fidelity while supporting precise event execution for time-sensitive operations and limited qubit lifetimes.
- 3.1 Quantum Network Simulation Requirements: The simulator separates reprogrammable functionality into modules, allowing alternative architectures, new protocols, applications, topologies, and traffic traces.
- 3.1 Quantum Network Simulation Requirements: Its stand-alone kernel supports large-scale photon-level discrete-event studies and portability to high-performance computing systems.
- 3.2 Modularized Design of SeQUeNCe: Because quantum-network architectures remain nonstandardized, SeQUeNCe makes minimal assumptions and uses five network modules plus a simulation kernel.
- 3.2 Modularized Design of SeQUeNCe: The simulation kernel stores timestamped events in a priority queue, executes the earliest event, and advances simulation time accordingly.
- 3.2 Modularized Design of SeQUeNCe: Hardware, entanglement, resource, network-management, and application modules collectively model devices, entanglement protocols, allocation, services, and user requests.
- 3.2 Modularized Design of SeQUeNCe: JSON configuration enables topology and parameter changes, while the modular design supports user-created hardware and protocol models.
- 3.2 Modularized Design of SeQUeNCe: An entanglement flow reserves resources, allocates memories, executes generation, purification, and swapping, and releases memories after application use.
4 DESIGN, IMPLEMENTATION, AND SIMULATION OF MODULES
SeQUeNCe implements a modular quantum-network simulator spanning hardware models, entanglement protocols, and network components. This section describes photonic hardware models and protocols for generating, purifying, and extending entanglement.
- Hardware and protocols: SeQUeNCe models quantum channels, classical channels, photon detectors, quantum memories, and entanglement-control protocols within a modular architecture.The hardware models expose interfaces for other modules to query and update state.
- Entanglement management: The Entanglement Management module implements Barrett-Kok generation, BBPSSW purification, and swapping to create, improve, and extend entanglement.Barrett-Kok uses two rounds and produces a maximally entangled memory pair, while swapping consumes shorter-distance pairs to connect distant memories.
- Hardware and protocols: Quantum channels schedule photon transmissions, model propagation delay and loss, and use time-division multiplexing to prevent simultaneous photons.Photon sources sharing a channel are synchronized to maintain proper spacing.
- Hardware and protocols: Single-photon detectors model efficiency, timestamp resolution, cooldown dead time, and Poisson-distributed dark counts.Detectors can also be used in Bell state measurement devices.
- Hardware and protocols: Quantum memories model single-atom qubits, excitation and expiration, photon-emission probability, and entanglement fidelity.For a stored state α|↓⟩ + β|↑⟩, the modeled emission probability is e|α|^2.
- Network management: Resource and network management coordinate local memory allocation, routing, reservation, and protocol instantiation across nodes.Router nodes provide full-stack functionality, while BSM nodes support Barrett-Kok generation on links.
5 THREE SIMULATION USE CASES
SeQUeNCe is used to study how hardware parameters, classical-control latency, and quantum-memory allocation affect a nine-node Chicago metropolitan quantum network. The simulations show that efficient memories, low-latency control, and load-aware memory redistribution improve network performance.
- Simulation Setup: The simulations model a nine-node, five-site Chicago topology augmented with three quantum links, with quantum routers, measured intersite delays, and repeated application requests.Each node has quantum memories and supports the protocols used in the simulations; requests run for 1,000 seconds.
- Comparison of Quantum Memory Parameters: Throughput generally improves with memory efficiency because higher-efficiency memories reduce entanglement-generation failures.Across the simulated cooperativities and efficiencies, increasing cooperativity provides less throughput improvement because fidelity is already high (>0.9).
- Impact of Classical Channel Delay: More than 10× higher average throughput is achieved for 20 kHz memories with low classical-channel delay, while long delays limit the benefit of increasing memory frequency from 2 kHz to 20 kHz.The simulations use e = 0.75 and C = 500 for this comparison.
- Two Memory Distribution Policies: 56.0% average memory usage at Argonne-3 and 56.6% at StarLight result from weighted redistribution, compared with 17.4% and 53.3% under even distribution.The weighted policy also reduces StarLight utilization above 90% from 43.5% to 24.4%.
- Two Memory Distribution Policies: 6.8% higher aggregate throughput and 69.1% more completed requests result from weighted memory distribution under the same total memory budget.The comparison repeats both policies ten times with different random seeds.
6 RELATED WORK
SeQUeNCe is situated among full-stack quantum-network simulators with different architectural assumptions and scopes. Existing tools vary in their support for physical, link, network, and application layers, as well as in the scale and scenarios they model.
- QuISP: QuISP supports entanglement purification, link tomography, and entanglement swapping, but published simulations focus on point-to-point scenarios up to 50 km.The cited work does not implement higher layers including routing and more complex network-topology studies.
- NetSquid: NetSquid work focuses on physical- and link-layer protocols and evaluates point-to-point communication over laboratory and intercity distances.The cited evaluations cover two nodes over 2 m and 25 km.
- Architectural Differences: SeQUeNCe uses cross-module communication in a modularized design, while NetSquid proposes five layers and QuISP distributes RuleSets along paths during connection setup.The three simulators differ in architectural assumptions and implementation details.
- Other Simulators: QuNetSim supports smaller-scale 5–10-node simulations and upper-layer modeling but leaves the physical layer unspecified and does not simulate repeater nodes.Other tools target individual quantum-network algorithms and protocols.
7 CONCLUSION
The paper introduces SeQUeNCe as a customizable discrete-event simulator with a modular architecture, high-performance simulation kernel, and broad quantum-network protocol support. It identifies additional protocols and parallelization as next steps while releasing the tool openly for community use and comparison.
- Contributions: SeQUeNCe combines a modularized quantum-network architecture with a high-performance kernel that tracks photon pulses and control messages with picosecond accuracy.The simulator is designed as a customizable discrete-event tool.
- Contributions: The simulator implements a comprehensive suite of quantum-network protocols as fully functional protocol state machines.This translates algorithmic protocol descriptions into executable simulations.
- Future Work: Additional protocols and parallelization for multinode, multicore supercomputers remain planned next steps.These additions are intended to support more comprehensive performance evaluations.
- Future Work: Open-source release is intended to encourage community use and extension and enable comparisons between experimental and simulation results.The stated goal is to generate community interest in extending SeQUeNCe.
A BARRETT-KOK PROTOCOL
The Barrett-Kok protocol generates entanglement through repeated excitation, photon detection, memory-state transformations, and detector-based state determination. SeQUeNCe represents the protocol with state machines distributed across quantum nodes and a Bell-state-measurement node, coordinated through classical messages.
- Protocol Steps: The protocol first excites both memories and waits for detector outcomes within a time window twait.Both-detector or no-detector outcomes indicate failure and require newly prepared qubits before retrying.
- Protocol Steps: During the relaxation interval trelax, memory states transform before both memories receive an X-gate and are excited again.The described sequence updates the joint memory state between detection attempts.
- Entanglement Outcome: A single detector click establishes a maximally entangled state, with the final Bell state determined by whether detections occur in the same or different detector.The protocol identifies the resulting state as |Ψ+⟩ or |Ψ−⟩.
- SeQUeNCe Implementation: SeQUeNCe instantiates Barrett-Kok state machines at quantum nodes and the BSM, beginning with a NEGOTIATE message and reporting results through an MR message.Entanglement lifetime begins at the first excite operation.
B BBPSSW PURIFICATION PROTOCOL
The BBPSSW protocol purifies two equal-fidelity entangled qubit pairs through local operations, measurement, and conditional retention. Its SeQUeNCe model determines success probabilistically from memory fidelity without updating memory states.
- BBPSSW uses two qubit pairs, A1B1 and A2B2, with equal fidelity F > 0.5.
- The protocol applies a local CNOT, measures A2 and B2, then discards A2B2.
- The purified pair A1B1 is retained when the measurement-result parity satisfies (ζ1 + ξ1)mod2 = 0.
- SeQUeNCe creates and pairs protocol instances through Resource Managers, then determines purification success probability from the two memories’ fidelity.
- The modeled purification randomly determines success using that probability but does not update the memories’ states.
C ENTANGLEMENT SWAPPING
The entanglement-swapping model separates intermediate-node operations from end-node responses. The intermediate node probabilistically determines success, communicates the result, and releases its resources, while end nodes update their entanglement identity after success.
- The swapping state machines distinguish roles at intermediate and end nodes.
- The intermediate node uses a success probability and random number to determine whether swapping succeeds.
- After the attempt, the intermediate node sends the outcome to paired end nodes and releases its resources.
- End nodes wait for the intermediate-node message and update the entanglement identity when swapping succeeds.
D RESERVATION PROTOCOL
The reservation protocol admits requests only when local memory is available throughout the requested interval, forwards accepted reservations along the path, and creates rules for generation, purification, and swapping.
- A reservation instance starts when a request arrives and checks local memory availability from the request’s start time through its end time.
- If sufficient memory is available, the instance attaches local information and forwards the request to the next hop until it reaches the Responder.
- Each quantum router along the path holds the reserved quantum-memory resources until the reservation is processed.
- The protocol defines service rules for entanglement generation, purification, and swapping.
- Raw reserved memory is allocated to generation, while equal-fidelity memories below the target fidelity are allocated to purification.