Quantum multi-programming for Grover's search

TL;DR

This paper introduces a quantum multi-programming algorithm for Grover's search that decomposes the algorithm and executes parts in parallel, resulting in higher success probability on noisy quantum computers.

Contribution

It proposes a novel quantum multi-programming approach for Grover's search that enhances success probability by decomposing and parallelizing the algorithm.

Findings

The new algorithm increases the success probability compared to canonical Grover's.

Empirical tests on IBM quantum computers validate the improved performance.

The approach effectively utilizes noisy intermediate-scale quantum computers.

Abstract

Quantum multi-programming is a method utilizing contemporary noisy intermediate-scale quantum computers by executing multiple quantum circuits concurrently. Despite early research on it, the research remains on quantum gates or small-size quantum algorithms without correlation. In this paper, we propose a quantum multi-programming (QMP) algorithm for Grover's search. Our algorithm decomposes Grover's algorithm by the partial diffusion operator and executes the decomposed circuits in parallel by QMP. We proved that this new algorithm increases the rotation angle of the Grover operator which, as a result, increases the success probability. The new algorithm is implemented on IBM quantum computers and compared with the canonical Grover's algorithm and other variations of Grover's algorithms. The empirical tests validate that our new algorithm outperforms other variations of Grover's algorithms as well as the canonical Grover's algorithm.