Quantum Computer Simulations at Warp Speed: Assessing the Impact of GPU Acceleration

TL;DR

This paper evaluates GPU acceleration for quantum computer simulations, demonstrating significant speedups with Nvidia GPUs and analyzing bottlenecks and optimizations in state vector simulators like Qiskit Aer.

Contribution

It provides a comprehensive benchmark and analysis of GPU and multi-GPU performance for quantum simulation, highlighting key functions and bottlenecks.

Findings

Up to 14x speedup with GPU over CPU for large qubit simulations

Nvidia cuQuantum offers 1.5-3x faster performance than Thrust backend

Data movement limits multi-GPU quantum simulation performance

Abstract

Quantum computer simulators are crucial for the development of quantum computing. In this work, we investigate the suitability and performance impact of GPU and multi-GPU systems on a widely used simulation tool - the state vector simulator Qiskit Aer. In particular, we evaluate the performance of both Qiskit's default Nvidia Thrust backend and the recent Nvidia cuQuantum backend on Nvidia A100 GPUs. We provide a benchmark suite of representative quantum applications for characterization. For simulations with a large number of qubits, the two GPU backends can provide up to 14x speedup over the CPU backend, with Nvidia cuQuantum providing further 1.5-3x speedup over the default Thrust backend. Our evaluation on a single GPU identifies the most important functions in Nvidia Thrust and cuQuantum for different quantum applications and their compute and memory bottlenecks. We also evaluate the gate fusion and cache-blocking optimizations on different quantum applications. Finally, we evaluate large-number qubit quantum applications on multi-GPU and identify data movement between host and GPU as the limiting factor for the performance.