FUFQ: A Heterogeneous-based Full-Amplitude Quantum Simulator

Quantum computing has unique advantages over classical computers in some aspects due to the superposition and entangled states of qubits. At present, due to factors such as expensive physical equipment, short coherence time, and harsh working conditions, using classical computers to perform quantum computing simulations has become an effective way to study quantum algorithms. However, some existing quantum simulators still have shortcomings such as slow calculation efficiency, inability to fully utilize the performance of multi-GPU devices, and poor portability. This paper proposes a full-amplitude quantum computing simulator FUFQ based on heterogeneous devices. Through the allocation of data and quantum gates and the use of GPU for calculation acceleration, in the simulation calculation of the test quantum circuit, compared with Qiskit and PyQpanda, the speed is increased by more than 2 times. And the quantum computing simulator proposed in this article is developed using C++ and CUDA, basically does not require additional libraries, and has good versatility and portability.