Virtual Network Allocation (SDN) in Cloud Computing Using Evolutionary Algorithm

• Published in: 2019 IEEE Latin American Conference on Computational Intelligence (LA-CCI) pp. 1 - 6
• Main Authors: Teixeira, Diego Andrade, de Amorim Monteiro, Thyago, da Silva Rodrigues, Ericks, de Queiroz Albuquerque, Edison, Fernandes, Sergio Murilo Maciel
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2019
• Subjects: cloud; Cloud computing; Delays; evolutionary algorithm; Evolutionary computation; Genetic algorithms; Resource management; SDN; Task analysis; Virtual machining
• DOI: 10.1109/LA-CCI47412.2019.9036766

Resource allocation is a key process for cloud networks, as Infrastructure as a Service (IaaS) and Software as a Service (SaaS). These technologies are becoming increasingly common, with increasing adoption by users and industries around the world. To make the most efficient use of this technology, we propose an optimized evolutionary algorithm (Evolution Strategy - ES) to achieve the optimization of the allocation of cloud resources on a physical network with large computational demands (data processing). We investigated the possibility of reallocating resources randomly generated on an SDN network and host them in a flexible and balanced manner. This improves performance and permit to find a better allocation for the maximum use of resources, processing and continuity of service. We consider that unbalanced allocation can lead to big problems in the network, such as a not full utilization of resources or even a standstill. The allocations made by the evolutionary algorithm (EA) proven to be more efficient than the allocations made by AGA (auto adaptive genetic algorithm) and conventional mode (random allocation). The experiments also show that our method improves the allocation of resources by giving more balance to the network.