Implementation of Ant Colony Optimization Combined with Tabu Search for Multi-resource Fair Allocation in Heterogeneous Cloud Computing

Resource allocation strategy has been a hot anddifficult research topic in the field of cloud computing. We address the problem of resource fairness allocation in heterogeneous cloud computing where the multiple types of resource are considered, which is computationally intractable. There is a significant gap between the solutions obtained by existing heuristic algorithmsand the optimal solutions, leading to lower resource utilization and unfair resource allocation. We propose a hybrid algorithm based on ant colony optimization (ACO) and Tabu Search (TS) to maximize the minimum global dominant share in heterogeneous servers. In order to balance the exploitation and exploration of the algorithm, the new self-adaptive parameter settings are introducedas uniformly random numbers to enhance the diversityof the population. Furthermore, we propose a revising operation to change infeasible solutions into feasible solutions. Compared with some algorithms from the literature, the experimental results indicate that our proposed algorithm can maximize the globaldominant share fairly and increase the resource utilization, and it is highly adaptable to different situations.