.png){kind=icon}
.png){kind=icon}
Return
QTest: A Simulation Testbed for Distributed Quantum Computing and Networking With Deep Reinforcement Learning
.png){kind=icon}
PRE
AI
DOI：10.1109/MIC.2025.3646208.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In distributed quantum computing (DQC), quantum networks interconnect multiple quantum processing units (QPUs) to overcome single-device limitations and enable cross-server computation. We present QTest, a modular simulation testbed for DQC and networking. The platform employs deep reinforcement learning-based quantum circuit partitioning and distributed quantum network scheduling, coordinating cross-server operations under network constraints such as routing and channel utilization. With containerized QPU nodes and GPU-accelerated back ends, QTest supports circuit execution across heterogeneous and geographically distributed environments. By enabling task partitioning, internode entanglement distribution, fidelity-aware routing, and latency-throughput evaluation, QTest provides a reproducible and extensible environment for protocol development and validation. The experimental results show that QTest achieves efficient runtime scalability under ideal conditions. Compared to CPU-based execution, GPU-accelerated distributed runs reduce execution time by up to an order of magnitude, validating the effectiveness of the proposed testbed.
Keywords:
Quantum computing
Logic gates
Qubit
Quantum entanglement
Routing
Artificial intelligence
Quantum networks
Quantum circuit
Processor scheduling
Internet