Swarm optimization algorithms applied to multi-resource fair allocation in heterogeneous cloud computing systems

Resource fair allocation is a challenging problem in heterogeneous cloud computing systems, both in real-life problems and for scientific research purposes. However, it is an NP-hard problem and solutions obtained by existing heuristic algorithms have a significant gap up to the optimal solutions. Motivated by this fact, we propose three swarm optimization algorithms: discrete artificial bee colony, discrete artificial fish swarm, and discrete shuffled frog leaping. In addition, we investigate how to utilize the impact of search behavior to improve the performance of the algorithm, and we design the self-adaptive parameter settings to balance between the exploitation and exploration of the algorithm. Furthermore, we propose a heuristic algorithm to generate a good initial solution. Compared with some algorithms from the literature, the simulation results show that our proposed algorithms can maximize the global dominant share fairly and increase the resource utilization, and they are highly adaptable to different situations.