Performance Analysis and Improvement of Parallel Differential Evolution

TL;DR

This paper analyzes the parallel computation of differential evolution (DE), introduces a new exponential crossover operator compatible with parallel execution, and demonstrates significant speed improvements in DE performance.

Contribution

It proposes a novel exponential crossover operator that enhances parallel computation efficiency in DE algorithms, leading to faster optimization.

Findings

The new crossover operator enables parallel execution with MKL/CUDA.

Parallel DE with the new operator significantly speeds up optimization.

The proposed structure outperforms traditional DE in computational speed.

Abstract

Differential evolution (DE) is an effective global evolutionary optimization algorithm using to solve global optimization problems mainly in a continuous domain. In this field, researchers pay more attention to improving the capability of DE to find better global solutions, however, the computational performance of DE is also a very interesting aspect especially when the problem scale is quite large. Firstly, this paper analyzes the design of parallel computation of DE which can easily be executed in Math Kernel Library (MKL) and Compute Unified Device Architecture (CUDA). Then the essence of the exponential crossover operator is described and we point out that it cannot be used for better parallel computation. Later, we propose a new exponential crossover operator (NEC) that can be executed parallelly with MKL/CUDA. Next, the extended experiments show that the new crossover operator can speed up DE greatly. In the end, we test the new parallel DE structure, illustrating that the former is much faster.