64-qubit quantum circuit simulation.

Area of Science:

• Quantum Computing
• Computational Physics
• High-Performance Computing

Background:

• Classical simulations of quantum circuits face limitations in space and time for systems exceeding 50 qubits.
• Quantum supremacy is demonstrated in systems with over 50 qubits, posing challenges for classical verification.
• Recent advancements by Google and IBM have explored simulation methods for larger quantum systems.

Purpose of the Study:

• To develop an efficient classical simulation scheme for large-scale quantum circuits.
• To enable the extraction of a significant number of measurement outcomes in a reduced timeframe.
• To provide a scalable and parallelizable approach for simulating quantum computations.

Main Methods:

• The proposed scheme utilizes a highly separable and parallelizable simulation process.
• It involves minimal inter-process communication, optimizing distributed computing.
• The method was tested on a 128-node cluster for a 64-qubit circuit and a single PC for smaller circuits.

Main Results:

• Successfully simulated a 64-qubit universal random circuit of depth 22 on a 128-node cluster.
• Achieved simulations of 56- and 42-qubit circuits on a single personal computer.
• Estimated simulation of a 72-qubit circuit (depth 23) in approximately 16 hours on a specified supercomputer.

Conclusions:

• The developed simulation scheme significantly reduces hardware burden for simulating more qubits.
• The parallelizable nature of the process allows for efficient extraction of measurement outcomes.
• This work offers a novel perspective and practical approach for classical simulations in the era of quantum supremacy.