Modified Elite Chaotic Immune Clonal Selection Algorithm for sever resource allocation in cloud computing systems

Cloud computing is a promising technology to improve computational efficiency for both IT enterprise and individuals. Resource allocation in cloud computing is very challenging as both server computing power and network bandwidth are limited. The computational efficiency of cloud computing system can be significantly improved if the resources are allocated in a balanced fashion. However, resource allocation in cloud computing is a multi-constrained nonlinear optimization problem. The computational complexity for an exhaustive search over all combinations of resource allocations is too high for practical implementation. In this paper, we develop a Modified Elite Chaotic Immune Clonal Selection Algorithm to increase the overall efficiency of the system. An elite strategy and chaotic approaches are designed to improve population diversity and escape from local optima. Performance comparisons are made with simulated annealing algorithm (SA) and three other heuristic algorithms. Simulation results show that the Modified Elite Chaotic Immune Clonal Selection Algorithm solves the resource allocation problem with higher system resource efficiency than all other heuristic algorithms.