A mechanism of auto elasticity based on response times for cloud computer enviroments and autossimilar workload

This paper proposes a cloud computing architecture based on containers and an algorithm for auto elasticity based on the response time requirements in a Web system. The proposed algorithm promotes an efficient allocation of containers to achieve expected response time in processing requests. This proposal was evaluated with a real time series obtained from a Web system hosted by Controladoria-Geral da União (CGU). The results show that the proposed algorithm achieves the expected response times allocating a lower number of containers than other related proposals.