Parallel Swarm Intelligence: Efficiency Study with Fast Range Search in Euclidean Space

Swarm intelligence algorithms are recognised for their effectiveness in solving complex optimisation problems. However, scalability can be a significant challenge, especially for large problem instances. This study focuses on examining the time performance of swarm intelligence algorithms when used with parallel computing on both central processing units (CPUs) and graphics processing units (GPUs). The investigation focuses on algorithms designed for range search in Euclidean space. The research optimizes these algorithms for GPU execution and assesses their effectiveness in handling large-scale instances. The inquiry also explores swarm-inspired solutions tailored for GPU implementation, with an emphasis on enhancing efficiency in video rendering and computer simulations. The research findings show potential for advancing the field by addressing challenges related to large-scale optimization through innovative GPU-accelerated swarm intelligence solutions.