作者: Diego de Freitas Bezerra , Guto Leoni Santos , Glauco Gonçalves , André Moreira , Leylane Graziele Ferreira da Silva
DOI: 10.1007/S11227-021-03620-Y
关键词: Task (project management) 、 Distributed computing 、 Mobile device 、 Virtual function 、 Resource allocation 、 Energy consumption 、 Cellular network 、 Computer science 、 Context (language use) 、 Resource (project management)
摘要: With the popularity of mobile devices, next generation networks has faced several challenges. Different applications have been emerged, with different requirements. Offering an infrastructure that meets types specific requirements is one these issues. In addition, due to user mobility, traffic generated by devices in a location not constant, making it difficult reach optimal resource allocation. this context, network function virtualization (NFV) can be used deploy telecommunication stacks as virtual functions running on commodity hardware meet users’ such performance and availability. However, deployment complex task. To select best placement strategy reduces usage, at same time keeps availability task, already proven NP-hard problem. Therefore, paper, we formulate NFV multi-objective problem, where risk associated energy consumption are taken into consideration. We propose usage two optimization algorithms, NSGA-II GDE3, solve These algorithms were consideration because both work problems present good performance. consider triathlon circuit scenario based real data from Ironman route use case evaluate compare algorithms. The results show GDE3 able attend objectives (minimize failure minimize consumption), while prioritizes consumption.